id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-08-22 00:00:00
| authors_parsed
listlengths 1
3.31k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.16654
|
Hubert Wu
|
Hubert Wu
|
The Empirical Welfare Content of International Price and Income Comparisons
| null | null | null | null |
econ.GN q-fin.EC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Multilateral index numbers, such as those used to make cross-country comparisons of prices and income, are fundamental objects in economics. However, these numbers can be challenging to interpret in terms of welfare due to a bias related to preference misspecification. To study this problem, I derive multilateral bounds on the cost of living implied by revealed preference. The bounds improve upon their classical bilateral counterparts, permit a benchmark-free appraisal of comparison methods, and define a novel index that is transitive, welfare-consistent, and which incorporates income effects. In an appraisal application I find support for the welfare-interpretability of many contemporary index methods, but not for market exchange rates. Turning to the end uses of cross-country comparisons themselves, I document that when calculated using an index based on the multilateral cost of living, the world in 2017 was substantially larger and more equal vis-\`a-vis the United States than suggested by conventional measures.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 12:22:12 GMT"
},
{
"version": "v2",
"created": "Fri, 23 May 2025 20:06:25 GMT"
},
{
"version": "v3",
"created": "Thu, 07 Aug 2025 10:14:39 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 21:00:47 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wu",
"Hubert",
""
]
] |
2504.16810
|
Richard Mandle
|
Jordan Hobbs, Calum J. Gibb, Richard J. Mandle
|
Polarity from the Bottom Up: A Computational Framework for Predicting Spontaneous Polar Order
| null | null |
10.1039/D5TC01641J
| null |
cond-mat.soft
|
http://creativecommons.org/licenses/by/4.0/
|
So-called polar liquid crystals possess spontaneous long-range mutual orientation of their electric dipole moments, conferring bulk polarity to fluid phases of matter. The combination of polarity and fluidity leads to complex phase behaviour, and rich new physics, yet the limited understanding around how specific molecular features generate long-range polar ordering in a fluid is a hindrance to development of new materials. In this work, we introduce a computational framework that probes the bimolecular potential energy landscape of candidate molecules, enabling us to dissect the role of directional intermolecular interactions in establishing polar order. In closely related families of materials we find conflicting preferences for (anti)parallel ordering which can be accounted for by specific interactions between molecules. Thus, our results allow us to argue that the presence (or absence) of polar order is a product of specific molecular features and strong directional intermolecular interactions rather than being simply a product of dipole-dipole forces. The design principles established can be leveraged to developing new polar liquid crystalline materials.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 15:28:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hobbs",
"Jordan",
""
],
[
"Gibb",
"Calum J.",
""
],
[
"Mandle",
"Richard J.",
""
]
] |
2504.16987
|
Shaoqing Ding
|
Shaoqing Ding, Jinyuan Yao, Zhen Bi, Quyen Tran, Bangzhi Liu, Qi Li, Susan Trolier-McKinstry, Thomas N. Jackson, Ying Liu
|
Indirect Tunneling Enabled Spontaneous Time-Reversal Symmetry Breaking and Josephson Diode Effect in TiN/Al$_2$O$_3$/Hf$_{0.8}$Zr$_{0.2}$O$_2$/Nb tunnel junctions
| null | null | null | null |
cond-mat.supr-con
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Josephson diode (JD) effect found in Josephson tunnel junctions (JTJs) has attracted a great deal of attention due to its importance for developing superconducting circuitry based quantum technologies. So far, the highly desirable electrical control of the JD effect has not been demonstrated in any JTJ prepared by techniques used in semiconductor industry. We report the fabrication of JTJs featuring a composite tunnel barrier of Al$_2$O$_3$ and Hf$_{\mathrm{0.8}}$Zr$_\mathrm{0.2}$O$_2$ prepared by complementary-metal-oxide-semiconductor (CMOS) compatible atomic layer deposition (ALD). These JTJs were found to show the JD effect in nominally zero magnetic fields with the nonreciprocity controllable using an electric training current, yielding a surprisingly large diode efficiency not achieved previously. The quasiparticle tunneling, through which the Josephson coupling in a JTJ is established, was found to show no nonreciprocity. We attribute these observations to the simultaneous presence of positive and negative Josephson couplings, with the latter originating from indirect tunneling. The resulted spontaneous time-reversal symmetry breaking and the double-minima washboard potential for the ensemble averaged phase difference in the resistively and capacitively shunted junction (RCSJ) model are shown to fully account for the experimentally observed JD effect.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 18:16:06 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ding",
"Shaoqing",
""
],
[
"Yao",
"Jinyuan",
""
],
[
"Bi",
"Zhen",
""
],
[
"Tran",
"Quyen",
""
],
[
"Liu",
"Bangzhi",
""
],
[
"Li",
"Qi",
""
],
[
"Trolier-McKinstry",
"Susan",
""
],
[
"Jackson",
"Thomas N.",
""
],
[
"Liu",
"Ying",
""
]
] |
2504.17444
|
Shushu Wu
|
Shushu Wu, Xiwei Wu, Qinxiang Cao
|
Encode the $\forall\exists$ Relational Hoare Logic into Standard Hoare Logic
|
Extended version of the paper accepted at OOPSLA 2025 R2
| null |
10.1145/3763138
| null |
cs.PL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Verifying a real-world program's functional correctness can be decomposed into (1) a refinement proof showing that the program implements a more abstract high-level program and (2) an algorithm correctness proof at the high level. Relational Hoare logic serves as a powerful tool to establish refinement but often necessitates formalization beyond standard Hoare logic. Particularly in the nondeterministic setting, the $\forall\exists$ relational Hoare logic is required. Existing approaches encode this logic into a Hoare logic with ghost states and invariants, yet these extensions significantly increase formalization complexity and soundness proof overhead. This paper proposes a generic encoding theory that reduces the $\forall\exists$ relational Hoare logic to standard (unary) Hoare logic. Precisely, we propose to redefine the validity of relational Hoare triples while preserving the original proof rules and then encapsulate the $\forall\exists$ pattern within assertions. We have proved that the validity of encoded standard Hoare triples is equivalent to the validity of the desired relational Hoare triples. Moreover, the encoding theory demonstrates how common relational Hoare logic proof rules are indeed special cases of standard Hoare logic proof rules, and relational proof steps correspond to standard proof steps. Our theory enables standard Hoare logic to prove $\forall\exists$ relational properties by defining a predicate Exec, without requiring modifications to the logic framework or re-verification of soundness.
|
[
{
"version": "v1",
"created": "Thu, 24 Apr 2025 11:13:54 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:06:36 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wu",
"Shushu",
""
],
[
"Wu",
"Xiwei",
""
],
[
"Cao",
"Qinxiang",
""
]
] |
2504.17838
|
Bernhard Jaeger
|
Bernhard Jaeger and Daniel Dauner and Jens Bei{\ss}wenger and Simon Gerstenecker and Kashyap Chitta and Andreas Geiger
|
CaRL: Learning Scalable Planning Policies with Simple Rewards
|
Accepted at the Conference on Robot Learning 2025
| null | null | null |
cs.LG cs.AI cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate reinforcement learning (RL) for privileged planning in autonomous driving. State-of-the-art approaches for this task are rule-based, but these methods do not scale to the long tail. RL, on the other hand, is scalable and does not suffer from compounding errors like imitation learning. Contemporary RL approaches for driving use complex shaped rewards that sum multiple individual rewards, \eg~progress, position, or orientation rewards. We show that PPO fails to optimize a popular version of these rewards when the mini-batch size is increased, which limits the scalability of these approaches. Instead, we propose a new reward design based primarily on optimizing a single intuitive reward term: route completion. Infractions are penalized by terminating the episode or multiplicatively reducing route completion. We find that PPO scales well with higher mini-batch sizes when trained with our simple reward, even improving performance. Training with large mini-batch sizes enables efficient scaling via distributed data parallelism. We scale PPO to 300M samples in CARLA and 500M samples in nuPlan with a single 8-GPU node. The resulting model achieves 64 DS on the CARLA longest6 v2 benchmark, outperforming other RL methods with more complex rewards by a large margin. Requiring only minimal adaptations from its use in CARLA, the same method is the best learning-based approach on nuPlan. It scores 91.3 in non-reactive and 90.6 in reactive traffic on the Val14 benchmark while being an order of magnitude faster than prior work.
|
[
{
"version": "v1",
"created": "Thu, 24 Apr 2025 17:56:01 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 16:46:42 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 19:00:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jaeger",
"Bernhard",
""
],
[
"Dauner",
"Daniel",
""
],
[
"Beißwenger",
"Jens",
""
],
[
"Gerstenecker",
"Simon",
""
],
[
"Chitta",
"Kashyap",
""
],
[
"Geiger",
"Andreas",
""
]
] |
2504.19062
|
Yu Zhang
|
Yu Zhang, Wenxiang Guo, Changhao Pan, Zhiyuan Zhu, Ruiqi Li, Jingyu Lu, Rongjie Huang, Ruiyuan Zhang, Zhiqing Hong, Ziyue Jiang, Zhou Zhao
|
Versatile Framework for Song Generation with Prompt-based Control
|
Accepted by Findings of EMNLP 2025
| null | null | null |
eess.AS cs.CL cs.SD
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Song generation focuses on producing controllable high-quality songs based on various prompts. However, existing methods struggle to generate vocals and accompaniments with prompt-based control and proper alignment. Additionally, they fall short in supporting various tasks. To address these challenges, we introduce VersBand, a multi-task song generation framework for synthesizing high-quality, aligned songs with prompt-based control. VersBand comprises these primary models: 1) VocalBand, a decoupled model, leverages the flow-matching method for generating singing styles, pitches, and mel-spectrograms, allowing fast, high-quality vocal generation with style control. 2) AccompBand, a flow-based transformer model, incorporates the Band-MOE, selecting suitable experts for enhanced quality, alignment, and control. This model allows for generating controllable, high-quality accompaniments aligned with vocals. 3) Two generation models, LyricBand for lyrics and MelodyBand for melodies, contribute to the comprehensive multi-task song generation system, allowing for extensive control based on multiple prompts. Experimental results show that VersBand outperforms baseline models across multiple song generation tasks using objective and subjective metrics. Demos and codes are available at https://aaronz345.github.io/VersBandDemo and https://github.com/AaronZ345/VersBand.
|
[
{
"version": "v1",
"created": "Sun, 27 Apr 2025 01:00:06 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Apr 2025 09:19:41 GMT"
},
{
"version": "v3",
"created": "Fri, 30 May 2025 10:15:41 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 05:29:52 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhang",
"Yu",
""
],
[
"Guo",
"Wenxiang",
""
],
[
"Pan",
"Changhao",
""
],
[
"Zhu",
"Zhiyuan",
""
],
[
"Li",
"Ruiqi",
""
],
[
"Lu",
"Jingyu",
""
],
[
"Huang",
"Rongjie",
""
],
[
"Zhang",
"Ruiyuan",
""
],
[
"Hong",
"Zhiqing",
""
],
[
"Jiang",
"Ziyue",
""
],
[
"Zhao",
"Zhou",
""
]
] |
2504.19326
|
Alex Ranne
|
Alex Ranne, Heiko Maier, Jinshi Zhao, Songli Moey, Ayhan Aktas, Manik Chana, Burak Temelkuran, Nassir Navab, Ferdinando Rodriguez Y Baena
|
Thermally Drawn Bioelectric Catheters: Enabling Proprioceptive Endovascular Navigation
|
23 pages, 8 figures
| null | null | null |
physics.med-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
To navigate medical instruments safely and accurately inside a patient's vascular tree, combining X-ray fluoroscopy with intermittent contrast injections is the gold standard. However, prolonged exposure to ionizing radiation poses health risks, necessitates the use of cumbersome lead vests for the clinicians, and contrast injections can lead to acute kidney injury in patients. Bioelectric Navigation, a non-fluoroscopic tracking modality, aims to provide an alternative. It uses weak electric currents to detect local anatomical features in the vasculature and localize instruments without x-ray imaging. In this work, we advance Bioelectric Navigation on two frontiers. Firstly, we introduce a new class of bespokely designed electrode catheters. They are fabricated using 3D printing, thermal drawing, and laser micro-machining. Specifically, we manufacture a 6 Fr catheter incorporating 16 electrodes, a guidewire channel and an additional sensor compartment. We thoroughly assess the catheter's mechanical and electrical properties. Secondly, we introduce an algorithm to localize the catheter along the centerline of a vascular phantom, for the first time fusing electric detection of vascular geometry with electric distance estimation. We report both tracking accuracy and usability evaluated by an expert endovascular surgeon, demonstrating the strong potential of this technology for integration into the existing clinical workflow.
|
[
{
"version": "v1",
"created": "Sun, 27 Apr 2025 18:52:01 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:05:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ranne",
"Alex",
""
],
[
"Maier",
"Heiko",
""
],
[
"Zhao",
"Jinshi",
""
],
[
"Moey",
"Songli",
""
],
[
"Aktas",
"Ayhan",
""
],
[
"Chana",
"Manik",
""
],
[
"Temelkuran",
"Burak",
""
],
[
"Navab",
"Nassir",
""
],
[
"Baena",
"Ferdinando Rodriguez Y",
""
]
] |
2504.19521
|
Itay Hazan
|
Itay Hazan, Idan Habler, Ron Bitton, Itsik Mantin
|
Security Steerability is All You Need
| null | null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
The adoption of Generative AI (GenAI) in applications inevitably comes with the expansion of the attack surface, combining new security threats along with the traditional ones. Consequently, numerous research and industrial initiatives aim to mitigate the GenAI related security threats by developing evaluation methods and designing defenses. However, while most of the GenAI security work focuses on universal threats (e.g. 'How to build a bomb'), there is significantly less discussion on application-level security and how to evaluate and mitigate it. Thus, in this work we adopt an application-centric approach to GenAI security, and show that while LLMs cannot protect against ad-hoc application specific threats, they can provide the framework for applications to protect themselves against such threats. Our first contribution is defining Security Steerability - a novel security measure for LLMs, assessing the model's capability to adhere to strict guardrails that are defined in the system prompt (e.g. 'Refrain from discussing about our competitors'). These guardrails, in case effective, can stop threats in the presence of malicious users who attempt to circumvent the application purpose. Our second contribution is a methodology to measure the security steerability of LLMs, utilizing a newly-developed benchmark called VeganRibs which assesses the LLM behavior in forcing specific guardrails that are not security per-se, in the presence of malicious user that tries to bypass the guardrails through prompt injection attacks with attack boosters (jailbreaks and perturbations). Using the new benchmark, we analyzed 18 open-source LLMs, demonstrating significant differences between their security steerability that are not trivial to foresee...
|
[
{
"version": "v1",
"created": "Mon, 28 Apr 2025 06:40:01 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Apr 2025 17:22:55 GMT"
},
{
"version": "v3",
"created": "Fri, 09 May 2025 20:10:47 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 07:12:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hazan",
"Itay",
""
],
[
"Habler",
"Idan",
""
],
[
"Bitton",
"Ron",
""
],
[
"Mantin",
"Itsik",
""
]
] |
2504.19675
|
Juho Inkinen
|
Osma Suominen, Juho Inkinen and Mona Lehtinen
|
Annif at SemEval-2025 Task 5: Traditional XMTC augmented by LLMs
|
6 pages, 4 figures, published at SemEval-2025 workshop Task 5: LLMs4Subjects: https://aclanthology.org/2025.semeval-1.315/
|
Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025), 2424--2431
| null | null |
cs.CL cs.AI cs.DL cs.IR cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
This paper presents the Annif system in SemEval-2025 Task 5 (LLMs4Subjects), which focussed on subject indexing using large language models (LLMs). The task required creating subject predictions for bibliographic records from the bilingual TIBKAT database using the GND subject vocabulary. Our approach combines traditional natural language processing and machine learning techniques implemented in the Annif toolkit with innovative LLM-based methods for translation and synthetic data generation, and merging predictions from monolingual models. The system ranked first in the all-subjects category and second in the tib-core-subjects category in the quantitative evaluation, and fourth in qualitative evaluations. These findings demonstrate the potential of combining traditional XMTC algorithms with modern LLM techniques to improve the accuracy and efficiency of subject indexing in multilingual contexts.
|
[
{
"version": "v1",
"created": "Mon, 28 Apr 2025 11:04:23 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:43:26 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Suominen",
"Osma",
""
],
[
"Inkinen",
"Juho",
""
],
[
"Lehtinen",
"Mona",
""
]
] |
2504.19793
|
Jiawen Shi
|
Jiawen Shi, Zenghui Yuan, Guiyao Tie, Pan Zhou, Neil Zhenqiang Gong, Lichao Sun
|
Prompt Injection Attack to Tool Selection in LLM Agents
| null | null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Tool selection is a key component of LLM agents. A popular approach follows a two-step process - \emph{retrieval} and \emph{selection} - to pick the most appropriate tool from a tool library for a given task. In this work, we introduce \textit{ToolHijacker}, a novel prompt injection attack targeting tool selection in no-box scenarios. ToolHijacker injects a malicious tool document into the tool library to manipulate the LLM agent's tool selection process, compelling it to consistently choose the attacker's malicious tool for an attacker-chosen target task. Specifically, we formulate the crafting of such tool documents as an optimization problem and propose a two-phase optimization strategy to solve it. Our extensive experimental evaluation shows that ToolHijacker is highly effective, significantly outperforming existing manual-based and automated prompt injection attacks when applied to tool selection. Moreover, we explore various defenses, including prevention-based defenses (StruQ and SecAlign) and detection-based defenses (known-answer detection, DataSentinel, perplexity detection, and perplexity windowed detection). Our experimental results indicate that these defenses are insufficient, highlighting the urgent need for developing new defense strategies.
|
[
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:36:43 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 15:39:44 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Shi",
"Jiawen",
""
],
[
"Yuan",
"Zenghui",
""
],
[
"Tie",
"Guiyao",
""
],
[
"Zhou",
"Pan",
""
],
[
"Gong",
"Neil Zhenqiang",
""
],
[
"Sun",
"Lichao",
""
]
] |
2504.19832
|
Dan Ben-Moshe
|
Dan Ben-Moshe, David Genesove
|
Assignment at the Frontier: Identifying the Frontier Structural Function and Bounding Mean Deviations
| null | null | null | null |
econ.EM econ.GN q-fin.EC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper analyzes a model in which an outcome equals a frontier function of inputs minus a nonnegative unobserved deviation. We allow the deviation's distribution to depend on inputs, thereby allowing for endogeneity. If zero lies in the support of the deviation given inputs -- an assumption we term assignment at the frontier -- then the frontier is identified by the supremum of the outcome at those inputs, obviating the need for instrumental variables. We then estimate the frontier, allowing for random error whose distribution may also depend on inputs. Finally, we derive a lower bound on the mean deviation, using only variance and skewness, that is robust to a scarcity of data near the frontier. We apply our methods to estimate a firm-level frontier production function and inefficiency.
|
[
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:34:54 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 14:12:37 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 15:19:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ben-Moshe",
"Dan",
""
],
[
"Genesove",
"David",
""
]
] |
2504.20791
|
Lukas H\"ormann
|
Lukas H\"ormann, Reinhard J. Maurer
|
Vibrational Energy Dissipation in Non-Contact Single-Molecule Junctions Governed by Local Geometry and Electronic Structure
| null | null | null | null |
cond-mat.mes-hall cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The vibrational dynamics of adsorbate molecules in single-molecule junctions depend critically on the geometric structure and electronic interactions between molecule and substrate. Vibrations, excited mechanochemically or by external stimuli, dissipate energy into substrate electrons and phonons. Energy dissipation leads to the broadening of spectral lines, vibrational lifetimes, and the coupling between molecular and substrate phonons. It affects molecular manipulation, giving rise to nanoscale friction, and contributes to scanning probe and surface spectroscopy signals. We present an approach to disentangle adsorbate vibrational dynamics in non-contact junctions by employing density functional theory, machine learning, and non-adiabatic molecular dynamics. Focusing on the CO-functionalised Cu surfaces representing a single-molecule junction, a widely studied system in scanning probe and energy dissipation experiments, we reveal strong vibrational mode specificity governed by the interplay of electron-phonon and phonon-phonon coupling. Electron-phonon relaxation rates vary by two orders of magnitude between modes and sensitively depend on the tip-substrate geometry. We find evidence of a weak non-additive effect between both energy dissipation channels, where electron-phonon coupling enhances phonon-phonon coupling. Our predicted vibrational lifetimes agree with infrared spectroscopy and helium scattering experiments. Finally, we outline how our findings can inform and enhance scanning probe experiments.
|
[
{
"version": "v1",
"created": "Tue, 29 Apr 2025 14:07:59 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 15:05:50 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hörmann",
"Lukas",
""
],
[
"Maurer",
"Reinhard J.",
""
]
] |
2504.21024
|
Tianqing Fang
|
Tianqing Fang, Hongming Zhang, Zhisong Zhang, Kaixin Ma, Wenhao Yu, Haitao Mi, Dong Yu
|
WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Model
|
EMNLP 2025 Main Conference
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Agent self-improvement, where the backbone Large Language Model (LLM) of the agent are trained on trajectories sampled autonomously based on their own policies, has emerged as a promising approach for enhancing performance. Recent advancements, particularly in web environments, face a critical limitation: their performance will reach a stagnation point during autonomous learning cycles, hindering further improvement. We argue that this stems from limited exploration of the web environment and insufficient exploitation of pre-trained web knowledge in LLMs. To improve the performance of self-improvement, we propose a novel framework that introduces a co-evolving World Model LLM. This world model predicts the next observation based on the current observation and action within the web environment. Leveraging LLMs' pretrained knowledge of abundant web content, the World Model serves dual roles: (1) as a virtual web server generating self-instructed training data to continuously refine the agent's policy, and (2) as an imagination engine during inference, enabling look-ahead simulation to guide action selection for the agent LLM. Experiments in real-world web environments (Mind2Web-Live, WebVoyager, and GAIA-web) show a 10% performance gain over existing self-evolving agents, demonstrating the efficacy and generalizability of our approach, without using any distillation from more powerful close-sourced models. Our work establishes the necessity of integrating world models into autonomous agent frameworks to unlock sustained adaptability. Code is available at https://github.com/Tencent/SelfEvolvingAgent
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 02:54:31 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:55:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fang",
"Tianqing",
""
],
[
"Zhang",
"Hongming",
""
],
[
"Zhang",
"Zhisong",
""
],
[
"Ma",
"Kaixin",
""
],
[
"Yu",
"Wenhao",
""
],
[
"Mi",
"Haitao",
""
],
[
"Yu",
"Dong",
""
]
] |
2504.21635
|
Sara Chrouf
|
Zeina Aldallal, Sara Chrouf, Khalil Hennara, Mohamed Motaism Hamed, Muhammad Hreden, Safwan AlModhayan
|
Sadeed: Advancing Arabic Diacritization Through Small Language Model
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Arabic text diacritization remains a persistent challenge in natural language processing due to the language's morphological richness. In this paper, we introduce Sadeed, a novel approach based on a fine-tuned decoder-only language model adapted from Kuwain 1.5B Hennara et al. [2025], a compact model originally trained on diverse Arabic corpora. Sadeed is fine-tuned on carefully curated, high-quality diacritized datasets, constructed through a rigorous data-cleaning and normalization pipeline. Despite utilizing modest computational resources, Sadeed achieves competitive results compared to proprietary large language models and outperforms traditional models trained on similar domains. Additionally, we highlight key limitations in current benchmarking practices for Arabic diacritization. To address these issues, we introduce SadeedDiac-25, a new benchmark designed to enable fairer and more comprehensive evaluation across diverse text genres and complexity levels. Together, Sadeed and SadeedDiac-25 provide a robust foundation for advancing Arabic NLP applications, including machine translation, text-to-speech, and language learning tools.
|
[
{
"version": "v1",
"created": "Wed, 30 Apr 2025 13:37:24 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:56:26 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Aldallal",
"Zeina",
""
],
[
"Chrouf",
"Sara",
""
],
[
"Hennara",
"Khalil",
""
],
[
"Hamed",
"Mohamed Motaism",
""
],
[
"Hreden",
"Muhammad",
""
],
[
"AlModhayan",
"Safwan",
""
]
] |
2505.01170
|
Meng Hua
|
Meng Hua, Chenghong Bian, Haotian Wu, Deniz G\"und\"uz
|
Realizing Fully-Connected Layers Over the Air via Reconfigurable Intelligent Surfaces
| null | null | null | null |
cs.IT eess.SP math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
By leveraging the waveform superposition property of the multiple access channel, over-the-air computation (AirComp) enables the execution of digital computations through analog means in the wireless domain, leading to faster processing and reduced latency. In this paper, we propose a novel approach to implement a neural network (NN) consisting of digital fully connected (FC) layers using physically reconfigurable hardware. Specifically, we investigate reconfigurable intelligent surfaces (RISs)-assisted multiple-input multiple-output (MIMO) systems to emulate the functionality of a NN for over-the-air inference. In this setup, both the RIS and the transceiver are jointly configured to manipulate the ambient wireless propagation environment, effectively reproducing the adjustable weights of a digital FC layer. We refer to this new computational paradigm as \textit{AirFC}. We formulate an imitation error minimization problem between the effective channel created by RIS and a target FC layer by jointly optimizing over-the-air parameters. To solve this non-convex optimization problem, an extremely low-complexity alternating optimization algorithm is proposed, where semi-closed-form/closed-form solutions for all optimization variables are derived. Simulation results show that the RIS-assisted MIMO-based AirFC can achieve competitive classification accuracy. Furthermore, it is also shown that a multi-RIS configuration significantly outperforms a single-RIS setup, particularly in line-of-sight (LoS)-dominated channels.
|
[
{
"version": "v1",
"created": "Fri, 02 May 2025 10:21:44 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 18:21:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hua",
"Meng",
""
],
[
"Bian",
"Chenghong",
""
],
[
"Wu",
"Haotian",
""
],
[
"Gündüz",
"Deniz",
""
]
] |
2505.01605
|
Eiji Konishi
|
Eiji Konishi
|
Information-acquiring von Neumann architecture of a computer: Functionality and subjectivity
|
7 pages, LaTeX, based on arXiv:1709.06719
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We design the information-acquiring von Neumann architecture of a computer in a fine-grained or coarse-grained model, where information is carried by classical bits. This architecture enables both a Hamiltonian process converting a given input pure state to another output pure state of the system to be considered (functionality) and a physical process to acquire information (subjectivity). The latter process is identified with the projection hypothesis in projective quantum measurement in the ensemble interpretation of quantum mechanics.
|
[
{
"version": "v1",
"created": "Fri, 02 May 2025 22:05:23 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 20:29:29 GMT"
},
{
"version": "v3",
"created": "Wed, 18 Jun 2025 06:53:13 GMT"
},
{
"version": "v4",
"created": "Tue, 22 Jul 2025 13:55:37 GMT"
},
{
"version": "v5",
"created": "Wed, 20 Aug 2025 22:27:35 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Konishi",
"Eiji",
""
]
] |
2505.01821
|
Jing Liu
|
Jing Liu, Yao Du, Kun Yang, Jiaqi Wu, Yan Wang, Xiping Hu, Zehua Wang, Yang Liu, Peng Sun, Azzedine Boukerche, Victor C.M. Leung
|
Edge-Cloud Collaborative Computing on Distributed Intelligence and Model Optimization: A Survey
|
43 pages, 10 figures, 10 tables
| null | null | null |
cs.DC cs.AI cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Edge-cloud collaborative computing (ECCC) has emerged as a pivotal paradigm for addressing the computational demands of modern intelligent applications, integrating cloud resources with edge devices to enable efficient, low-latency processing. Recent advancements in AI, particularly deep learning and large language models (LLMs), have dramatically enhanced the capabilities of these distributed systems, yet introduce significant challenges in model deployment and resource management. In this survey, we comprehensive examine the intersection of distributed intelligence and model optimization within edge-cloud environments, providing a structured tutorial on fundamental architectures, enabling technologies, and emerging applications. Additionally, we systematically analyze model optimization approaches, including compression, adaptation, and neural architecture search, alongside AI-driven resource management strategies that balance performance, energy efficiency, and latency requirements. We further explore critical aspects of privacy protection and security enhancement within ECCC systems and examines practical deployments through diverse applications, spanning autonomous driving, healthcare, and industrial automation. Performance analysis and benchmarking techniques are also thoroughly explored to establish evaluation standards for these complex systems. Furthermore, the review identifies critical research directions including LLMs deployment, 6G integration, neuromorphic computing, and quantum computing, offering a roadmap for addressing persistent challenges in heterogeneity management, real-time processing, and scalability. By bridging theoretical advancements and practical deployments, this survey offers researchers and practitioners a holistic perspective on leveraging AI to optimize distributed computing environments, fostering innovation in next-generation intelligent systems.
|
[
{
"version": "v1",
"created": "Sat, 03 May 2025 13:55:38 GMT"
},
{
"version": "v2",
"created": "Sat, 17 May 2025 11:18:55 GMT"
},
{
"version": "v3",
"created": "Tue, 12 Aug 2025 16:02:52 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 03:51:19 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Liu",
"Jing",
""
],
[
"Du",
"Yao",
""
],
[
"Yang",
"Kun",
""
],
[
"Wu",
"Jiaqi",
""
],
[
"Wang",
"Yan",
""
],
[
"Hu",
"Xiping",
""
],
[
"Wang",
"Zehua",
""
],
[
"Liu",
"Yang",
""
],
[
"Sun",
"Peng",
""
],
[
"Boukerche",
"Azzedine",
""
],
[
"Leung",
"Victor C. M.",
""
]
] |
2505.01897
|
Miroslav Maksimovi\'c
|
Miroslav D. Maksimovi\'c, Milan Lj. Zlatanovi\'c and Milica R. Vu\v{c}urovi\'c
|
Lorentzian manifolds equipped with a concircularly semi-symmetric metric connection
|
In accordance with the reviewer's comments (for the Journal of Geometry and Physics), we have added an explanation for the mutual connection and its geometric interpretation, as well as Remark 5.1
| null | null | null |
math.DG gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Building upon previous works characterizing GRW space-times using concircular and torse-forming vectors, this paper investigates a Lorentzian manifold equipped with a concircularly semi-symmetric metric connection. We demonstrate that such a manifold reduces to a GRW space-time under specific conditions: when the generator of the observed connection is a unit timelike vector. Also, in that case, the mentioned connection becomes a semi-symmetric metric $P$-connection. The non-zero nature of the three curvature tensors and their corresponding Ricci tensors motivates an exploration of manifold symmetries. In this way, we derive necessary and sufficient conditions for the manifold to be Einstein and we prove that a perfect fluid space-time with a semi-symmetric metric $P$-connection is Ricci pseudo-symmetric manifold of constant type. Furthermore, we show that if this space-time satisfies the Einstein's field equations without the cosmological constant, the strong energy condition is violated.
|
[
{
"version": "v1",
"created": "Sat, 03 May 2025 19:10:02 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 13:23:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Maksimović",
"Miroslav D.",
""
],
[
"Zlatanović",
"Milan Lj.",
""
],
[
"Vučurović",
"Milica R.",
""
]
] |
2505.01962
|
Haochuan Wang
|
Haochuan Wang
|
Heterogeneous Trader Responses to Macroeconomic Surprises: Simulating Order Flow Dynamics
|
9 pages, 5 pages, with dgp code attached
| null | null | null |
econ.GN q-fin.EC q-fin.PM
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Understanding how market participants react to shocks like scheduled macroeconomic news is crucial for both traders and policymakers. We develop a calibrated data generation process DGP that embeds four stylized trader archetypes retail, pension, institutional, and hedge funds into an extended CAPM augmented by CPI surprises. Each agents order size choice is driven by a softmax discrete choice rule over small, medium, and large trades, where utility depends on risk aversion, surprise magnitude, and liquidity. We aim to analyze each agent's reaction to shocks and Monte Carlo experiments show that higher information, lower aversion agents take systematically larger positions and achieve higher average wealth. Retail investors under react on average, exhibiting smaller allocations and more dispersed outcomes. And ambient liquidity amplifies the sensitivity of order flow to surprise shocks. Our framework offers a transparent benchmark for analyzing order flow dynamics around macro releases and suggests how real time flow data could inform news impact inference.
|
[
{
"version": "v1",
"created": "Sun, 04 May 2025 02:12:53 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:48:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Haochuan",
""
]
] |
2505.03898
|
Ying Yuan
|
Shuqi Wang, Ying Yuan, Suyu Liu
|
ROSE: Randomized Optimal Selection Design for Dose Optimization
| null | null | null | null |
stat.ME
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The U.S. Food and Drug Administration (FDA) launched Project Optimus to shift the objective of dose selection from the maximum tolerated dose to the optimal biological dose (OBD), optimizing the benefit-risk tradeoff. One approach recommended by the FDA's guidance is to conduct randomized trials comparing multiple doses. In this paper, using the selection design framework (Simon et al., 1985), we propose a randomized optimal selection (ROSE) design, which minimizes sample size while ensuring the probability of correct selection of the OBD at prespecified accuracy levels. The ROSE design is simple to implement, involving a straightforward comparison of the difference in response rates between two dose arms against a predetermined decision boundary. We further consider a two-stage ROSE design that allows for early selection of the OBD at the interim when there is sufficient evidence, further reducing the sample size. Simulation studies demonstrate that the ROSE design exhibits desirable operating characteristics in correctly identifying the OBD. A sample size of 15 to 40 patients per dosage arm typically results in a percentage of correct selection of the optimal dose ranging from 60% to 70%.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 18:09:26 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 18:37:04 GMT"
},
{
"version": "v3",
"created": "Fri, 13 Jun 2025 16:05:03 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 02:43:55 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Shuqi",
""
],
[
"Yuan",
"Ying",
""
],
[
"Liu",
"Suyu",
""
]
] |
2505.03954
|
Matthew Kwan
|
Vishesh Jain, Matthew Kwan, Dhruv Mubayi, Tuan Tran
|
The edge-statistics conjecture for hypergraphs
| null | null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
Let $r,k,\ell$ be integers such that $0\le\ell\le\binom{k}{r}$. Given a large $r$-uniform hypergraph $G$, we consider the fraction of $k$-vertex subsets which span exactly $\ell$ edges. If $\ell$ is 0 or $\binom{k}{r}$, this fraction can be exactly 1 (by taking $G$ to be empty or complete), but for all other values of $\ell$, one might suspect that this fraction is always significantly smaller than 1.
In this paper we prove an essentially optimal result along these lines: if $\ell$ is not 0 or $\binom{k}{r}$, then this fraction is at most $(1/e) + \varepsilon$, assuming $k$ is sufficiently large in terms of $r$ and $\varepsilon>0$, and $G$ is sufficiently large in terms of $k$. Previously, this was only known for a very limited range of values of $r,k,\ell$ (due to Kwan-Sudakov-Tran, Fox-Sauermann, and Martinsson-Mousset-Noever-Truji\'{c}). Our result answers a question of Alon-Hefetz-Krivelevich-Tyomkyn, who suggested this as a hypergraph generalisation of their "edge-statistics conjecture". We also prove a much stronger bound when $\ell$ is far from 0 and $\binom{k}{r}$.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 20:14:09 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 08:31:30 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jain",
"Vishesh",
""
],
[
"Kwan",
"Matthew",
""
],
[
"Mubayi",
"Dhruv",
""
],
[
"Tran",
"Tuan",
""
]
] |
2505.04612
|
Jiahao Li
|
Jiahao Li, Haochen Wang, Muhammad Zubair Irshad, Igor Vasiljevic, Matthew R. Walter, Vitor Campagnolo Guizilini, Greg Shakhnarovich
|
FastMap: Revisiting Structure from Motion through First-Order Optimization
|
Project webpage: https://jiahao.ai/fastmap
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We propose FastMap, a new global structure from motion method focused on speed and simplicity. Previous methods like COLMAP and GLOMAP are able to estimate high-precision camera poses, but suffer from poor scalability when the number of matched keypoint pairs becomes large, mainly due to the time-consuming process of second-order Gauss-Newton optimization. Instead, we design our method solely based on first-order optimizers. To obtain maximal speedup, we identify and eliminate two key performance bottlenecks: computational complexity and the kernel implementation of each optimization step. Through extensive experiments, we show that FastMap is up to 10 times faster than COLMAP and GLOMAP with GPU acceleration and achieves comparable pose accuracy.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 17:56:15 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 03:10:38 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 01:12:14 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Li",
"Jiahao",
""
],
[
"Wang",
"Haochen",
""
],
[
"Irshad",
"Muhammad Zubair",
""
],
[
"Vasiljevic",
"Igor",
""
],
[
"Walter",
"Matthew R.",
""
],
[
"Guizilini",
"Vitor Campagnolo",
""
],
[
"Shakhnarovich",
"Greg",
""
]
] |
2505.06521
|
Kin-Ya Oda
|
Satoshi Mishima and Kin-ya Oda
|
Implementing Errors on Errors: Bayesian vs Frequentist
|
Version accepted for publication in Eur. Phys. J. C; footnotes 2, 5, and 11 added; Refs. [13,14] added; minor revisions; 11 pages, 1 figure
| null | null | null |
hep-ph astro-ph.IM hep-ex stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
When combining apparently inconsistent experimental results, one often implements errors on errors. The Particle Data Group's phenomenological prescription offers a practical solution but lacks a firm theoretical foundation. To address this, D'Agostini and Cowan have proposed Bayesian and frequentist approaches, respectively, both introducing gamma-distributed auxiliary variables to model uncertainty in quoted errors. In this Letter, we show that these two formulations admit a parameter-by-parameter correspondence, and are structurally equivalent. This identification clarifies how Bayesian prior choices can be interpreted in terms of frequentist sampling assumptions, providing a unified probabilistic framework for modeling uncertainty in quoted variances.
|
[
{
"version": "v1",
"created": "Sat, 10 May 2025 05:50:32 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 00:54:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mishima",
"Satoshi",
""
],
[
"Oda",
"Kin-ya",
""
]
] |
2505.06911
|
Lishan Yang
|
Lishan Yang, Wei Emma Zhang, Quan Z. Sheng, Lina Yao, Weitong Chen and Ali Shakeri
|
MMiC: Mitigating Modality Incompleteness in Clustered Federated Learning
|
9 pages
| null |
10.1145/3746252.3761140
| null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by-sa/4.0/
|
In the era of big data, data mining has become indispensable for uncovering hidden patterns and insights from vast and complex datasets. The integration of multimodal data sources further enhances its potential. Multimodal Federated Learning (MFL) is a distributed approach that enhances the efficiency and quality of multimodal learning, ensuring collaborative work and privacy protection. However, missing modalities pose a significant challenge in MFL, often due to data quality issues or privacy policies across the clients. In this work, we present MMiC, a framework for Mitigating Modality incompleteness in MFL within the Clusters. MMiC replaces partial parameters within client models inside clusters to mitigate the impact of missing modalities. Furthermore, it leverages the Banzhaf Power Index to optimize client selection under these conditions. Finally, MMiC employs an innovative approach to dynamically control global aggregation by utilizing Markovitz Portfolio Optimization. Extensive experiments demonstrate that MMiC consistently outperforms existing federated learning architectures in both global and personalized performance on multimodal datasets with missing modalities, confirming the effectiveness of our proposed solution. Our code is available at https://github.com/gotobcn8/MMiC.
|
[
{
"version": "v1",
"created": "Sun, 11 May 2025 09:12:36 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 13:11:28 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 10:38:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yang",
"Lishan",
""
],
[
"Zhang",
"Wei Emma",
""
],
[
"Sheng",
"Quan Z.",
""
],
[
"Yao",
"Lina",
""
],
[
"Chen",
"Weitong",
""
],
[
"Shakeri",
"Ali",
""
]
] |
2505.06936
|
Mohammad Mashayekhi -
|
Mohammad Mashayekhi and Kamran Salehian
|
AI-Powered Inverse Design of Ku-Band SIW Resonant Structures by Iterative Residual Correction Network
|
18 pages, 14 figures
|
AEUE - International Journal of Electronics and Communications, 2025
|
10.1016/j.aeue.2025.156003
|
AEUE 156003
|
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Inverse electromagnetic modeling has emerged as a powerful approach for designing complex microwave structures with high accuracy and efficiency. In this study, we propose an Iterative Residual Correction Network (IRC-Net) for the inverse design of Ku-band Substrate Integrated Waveguide (SIW) components based on multimode resonators. We use a multimode resonance structure to demonstrate that it is possible to control the resonances of the structure. Therefore, these structures can be used for resonant components and smart filter design. The proposed deep learning architecture leverages residual neural networks to overcome the limitations of traditional inverse design techniques, such as the Feedforward Inverse Model (FIM), offering improved generalization and prediction accuracy. The approach begins with a FIM to generate initial design estimates, followed by an iterative correction strategy inspired by the Hybrid Inverse-Forward Residual Refinement Network (HiFR\textsuperscript{2}-Net), which we call IRC-Net. Experiments demonstrate that the IRC-Net achieves substantial improvements in prediction accuracy compared to traditional single-stage networks, validated through statistical metrics, full-wave electromagnetic simulations, and measurements. To validate the proposed framework, we first design and fabricate a three-resonance SIW structure. Next, we apply the trained IRC-Net model to predict the geometry of a four-resonance structure based on its desired frequency response. Both designs are fabricated and tested, showing strong agreement between the simulated, predicted, and measured results, confirming the effectiveness and practicality of the proposed method.
|
[
{
"version": "v1",
"created": "Sun, 11 May 2025 10:51:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mashayekhi",
"Mohammad",
""
],
[
"Salehian",
"Kamran",
""
]
] |
2505.07217
|
A. K. M. Selim Reza
|
A.K.M. Selim Reza, Manabu Oura, Masashi Kosuda, Shoyu Nagaoka
|
Computation of Vector-Valued Invariants for a Finite Complex Reflection Group
| null | null | null | null |
math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We consider the complex reflection group \( \mathcal{G} \), identified as No. 8 in the Shephard-Todd classification. In this paper, we present computations of the vector-valued invariants associated with various representations of \( \mathcal{G} \). Additionally, we investigate the structure of the corresponding invariant rings.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 04:12:16 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 02:18:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Reza",
"A. K. M. Selim",
""
],
[
"Oura",
"Manabu",
""
],
[
"Kosuda",
"Masashi",
""
],
[
"Nagaoka",
"Shoyu",
""
]
] |
2505.07350
|
Philipp Reineck
|
Roy Styles, Mengke Han, Toon Goris, James Partridge, Brett C. Johnson, Blanca del Rosal, Amanda N. Abraham, Heike Ebendorff-Heidepriem, Brant C. Gibson, Nikolai Dontschuk, Jean-Philippe Tetienne, Philipp Reineck
|
All-optical electric field sensing with nanodiamond-doped polymer thin films
| null | null |
10.1002/adfm.202512068
| null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The nitrogen-vacancy (NV) center is a photoluminescent defect in diamond that exists in different charge states, NV$^-$ and NV$^0$, that are sensitive to the NV's nanoscale environment. Here, we show that photoluminescence (PL) from NV centers in fluorescent nanodiamonds (FNDs) can be employed for all-optical voltage sensing based on electric field-induced NV charge state modulation. More than 95% of FNDs integrated into a capacitor device show a transient increase in NV$^-$ PL intensity of up to 31% within 0.1 ms after application of an external voltage, accompanied by a simultaneous decrease in NV$^0$ PL. The change in NV$^-$ PL increases with increasing applied voltage from 0 to 100 V, corresponding to an electric field of 0 to 625 kV cm$^ {-1}$ in our devices. The electric field sensitivity of a single FND is 19 V cm$^{-1}$ Hz$^ {-1/2}$. We investigate the NV charge state photodynamics on the millisecond timescale and find that the change in NV PL strongly depends on the rate of photoexcitation. We propose a model that qualitatively explains the observed changes in NV PL based on an electric field-induced redistribution of photoexcited electrons from substitutional nitrogen defects to NV centers, leading to a transient conversion of NV$^0$ to NV$^-$ centers upon application of an external voltage. Our results contribute to the development of FNDs as reliable, all-optical, nanoscale electric field sensors in solid-state systems.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 08:44:10 GMT"
},
{
"version": "v2",
"created": "Tue, 17 Jun 2025 00:26:21 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Styles",
"Roy",
""
],
[
"Han",
"Mengke",
""
],
[
"Goris",
"Toon",
""
],
[
"Partridge",
"James",
""
],
[
"Johnson",
"Brett C.",
""
],
[
"del Rosal",
"Blanca",
""
],
[
"Abraham",
"Amanda N.",
""
],
[
"Ebendorff-Heidepriem",
"Heike",
""
],
[
"Gibson",
"Brant C.",
""
],
[
"Dontschuk",
"Nikolai",
""
],
[
"Tetienne",
"Jean-Philippe",
""
],
[
"Reineck",
"Philipp",
""
]
] |
2505.07719
|
Hyunwoo Oh
|
Hyunwoo Oh
|
Training neural control variates using correlated configurations
|
9 pages, 8 figures; Version accepted by PRD
| null | null | null |
hep-lat cs.LG nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Neural control variates (NCVs) have emerged as a powerful tool for variance reduction in Monte Carlo (MC) simulations, particularly in high-dimensional problems where traditional control variates are difficult to construct analytically. By training neural networks to learn auxiliary functions correlated with the target observable, NCVs can significantly reduce estimator variance while preserving unbiasedness. However, a critical but often overlooked aspect of NCV training is the role of autocorrelated samples generated by Markov Chain Monte Carlo (MCMC). While such samples are typically discarded for error estimation due to their statistical redundancy, they may contain useful information about the structure of the underlying probability distribution that can benefit the training process. In this work, we systematically examine the effect of using correlated configurations in training neural control variates. We demonstrate, both conceptually and numerically, that training on correlated data can improve control variate performance, especially in settings with limited computational resources. Our analysis includes empirical results from $U(1)$ gauge theory and scalar field theory, illustrating when and how autocorrelated samples enhance NCV construction. These findings provide practical guidance for the efficient use of MCMC data in training neural networks.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 16:25:00 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 17:23:43 GMT"
},
{
"version": "v3",
"created": "Tue, 15 Jul 2025 16:34:42 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 16:40:24 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Oh",
"Hyunwoo",
""
]
] |
2505.07767
|
Pavel Alekseev
|
P. S. Alekseev and M. A. Semina
|
Analytical model for non-linear magnetotransport in viscous electron fluid
|
13 pages, 3 figures
| null | null | null |
cond-mat.mes-hall cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We develop an analytical theoretical model for non-linear hydrodynamic magnetotransport of two-dimensional (2D) electron fluid with strong pair correlations in the electron dynamics. Within classical kinetics of 2D electrons, such correlations are described as subsequent ``extended'' collisions of the same electrons, temporarily joined in pairs. Corresponding correlation-induced retarded terms in the fluid dynamic equations can be described for slow flows as the dependence of the electron fluid viscosity on the flow velocity gradient, that is Non-Newtonian behavior of the fluid. We analytically calculate flow profiles in long samples in a stationary highly non-linear regime and the corresponding magnetoresistance. Pair correlations lead to a characteristic non-monotonic dependence of the differential resistance on magnetic field. We compare our results with experimental data on non-linear magnetotransport of high-purity GaAs quantum wells; we conclude that our model can be responsible for a part of the observed features of the differential magnetoresistance.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 17:18:18 GMT"
},
{
"version": "v2",
"created": "Tue, 13 May 2025 16:58:15 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 05:35:13 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Alekseev",
"P. S.",
""
],
[
"Semina",
"M. A.",
""
]
] |
2505.08093
|
Charles Wade
|
Charles Wade, Devon Beck, Robert MacCurdy
|
Implicit Toolpath Generation for Functionally Graded Additive Manufacturing via Gradient-Informed Slicing
| null | null | null | null |
cs.CG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper presents a novel gradient-informed slicing method for functionally graded additive manufacturing (FGM) that overcomes the limitations of conventional toolpath planning approaches, which struggle to produce truly continuous gradients. By integrating multi-material gradients into the toolpath generation process, our method enables the fabrication of FGMs with complex gradients that vary seamlessly in any direction. We leverage OpenVCAD's implicit representation of geometry and material fields to directly extract iso-contours, enabling accurate, controlled gradient toolpaths. Two novel strategies are introduced to integrate these gradients into the toolpath planning process. The first strategy maintains traditional perimeter, skin, and infill structures subdivided by mixture ratios, with automated 'zippering' to mitigate stress concentrations. The second strategy fills iso-contoured regions densely, printing directly against gradients to eliminate purging and reduce waste. Both strategies accommodate gradually changing printing parameters, such as mixed filament ratios, toolhead switching, and variable nozzle temperatures for foaming materials. This capability allows for controlled variation of composition, density, and other properties within a single build, expanding the design space for functionally graded parts. Experimental results demonstrate the fabrication of high-quality FGMs with complex, multi-axis gradients, highlighting the versatility of our method. We showcase the successful implementation of both strategies on a range of geometries and material combinations, demonstrating the potential of our approach to produce intricate and functional FGMs. This work provides a robust, open-source, and automated framework for designing and fabricating advanced FGMs, accelerating research in multi-material additive manufacturing.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 21:56:32 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 15:26:53 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wade",
"Charles",
""
],
[
"Beck",
"Devon",
""
],
[
"MacCurdy",
"Robert",
""
]
] |
2505.09193
|
Wei Jiang
|
Wei Jiang, Junru Li, Kai Zhang, Li Zhang
|
BiECVC: Gated Diversification of Bidirectional Contexts for Learned Video Compression
|
Accepted to ACMMM 2025
| null |
10.1145/3746027.3754725
| null |
eess.IV cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent forward prediction-based learned video compression (LVC) methods have achieved impressive results, even surpassing VVC reference software VTM under the Low Delay B (LDB) configuration. In contrast, learned bidirectional video compression (BVC) remains underexplored and still lags behind its forward-only counterparts. This performance gap is mainly due to the limited ability to extract diverse and accurate contexts: most existing BVCs primarily exploit temporal motion while neglecting non-local correlations across frames. Moreover, they lack the adaptability to dynamically suppress harmful contexts arising from fast motion or occlusion. To tackle these challenges, we propose BiECVC, a BVC framework that incorporates diversified local and non-local context modeling along with adaptive context gating. For local context enhancement, BiECVC reuses high-quality features from lower layers and aligns them using decoded motion vectors without introducing extra motion overhead. To model non-local dependencies efficiently, we adopt a linear attention mechanism that balances performance and complexity. To further mitigate the impact of inaccurate context prediction, we introduce Bidirectional Context Gating, inspired by data-dependent decay in recent autoregressive language models, to dynamically filter contextual information based on conditional coding results. Extensive experiments demonstrate that BiECVC achieves state-of-the-art performance, reducing the bit-rate by 13.4% and 15.7% compared to VTM 13.2 under the Random Access (RA) configuration with intra periods of 32 and 64, respectively. To our knowledge, BiECVC is the first learned video codec to surpass VTM 13.2 RA across all standard test datasets.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 06:55:37 GMT"
},
{
"version": "v2",
"created": "Thu, 15 May 2025 01:32:30 GMT"
},
{
"version": "v3",
"created": "Sun, 06 Jul 2025 08:47:36 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Jul 2025 16:57:30 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jiang",
"Wei",
""
],
[
"Li",
"Junru",
""
],
[
"Zhang",
"Kai",
""
],
[
"Zhang",
"Li",
""
]
] |
2505.09452
|
Umut Tanyeri
|
Umut Tanyeri, Ahmed Kallushi, R{\i}fat Onur Umucal{\i}lar and Ahmet Kele\c{s}
|
Vortex and fractional quantum Hall phases in a rotating anisotropic Bose gas
|
10 pages, 4 figures
|
Physical Review A 112, 023321 (2025)
|
10.1103/bsvm-vgnb
| null |
cond-mat.quant-gas
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Realizing fractional quantum Hall (FQH) states in ultracold atomic systems remains a major goal despite numerous experimental advances in the last few decades. Recent progress in trap anisotropy control under rapid rotation has renewed interest in ultracold atomic FQH physics, enabling experiments that impart much larger angular momentum per particle and offer in-situ imaging with resolution finer than the cyclotron orbit size. In this paper, we present a theoretical investigation of a rapidly rotating anisotropic Bose gas. By projecting the full Hamiltonian, including both kinetic and interaction terms, onto the lowest Landau level, we derive a compact two-parameter model that captures the effects of interaction strength, rotation rate, and anisotropy. Using exact diagonalization and density matrix renormalization group, we obtain a phase diagram that features broken-symmetry phases and topologically ordered quantum Hall states, while also highlighting the distinctive physics arising from the system's edges. Our results demonstrate the potential for future theoretical and experimental exploration of anisotropic quantum fluids, offering a unified framework for weakly interacting Bose condensates, vortex matter, and strongly correlated topological phases.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 15:03:21 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:20:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Tanyeri",
"Umut",
""
],
[
"Kallushi",
"Ahmed",
""
],
[
"Umucalılar",
"Rıfat Onur",
""
],
[
"Keleş",
"Ahmet",
""
]
] |
2505.11378
|
Alexander Kim
|
Alexander Kim, Charlotte Botha
|
Machine Learning Approaches to Vocal Register Classification in Contemporary Male Pop Music
|
8 pages, 8 figures
| null | null | null |
cs.SD cs.LG eess.AS
|
http://creativecommons.org/licenses/by/4.0/
|
For singers of all experience levels, one of the most daunting challenges in learning technical repertoire is navigating placement and vocal register in and around the passagio (passage between chest voice and head voice registers). Particularly in pop music, where a single artist may use a variety of timbre's and textures to achieve a desired quality, it can be difficult to identify what vocal register within the vocal range a singer is using. This paper presents two methods for classifying vocal registers in an audio signal of male pop music through the analysis of textural features of mel-spectrogram images. Additionally, we will discuss the practical integration of these models for vocal analysis tools, and introduce a concurrently developed software called AVRA which stands for Automatic Vocal Register Analysis. Our proposed methods achieved consistent classification of vocal register through both Support Vector Machine (SVM) and Convolutional Neural Network (CNN) models, which supports the promise of more robust classification possibilities across more voice types and genres of singing.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 15:41:28 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:44:02 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Kim",
"Alexander",
""
],
[
"Botha",
"Charlotte",
""
]
] |
2505.11517
|
Daniel T. Speckhard
|
Daniel T. Speckhard
|
Graph topology estimation of power grids using pairwise mutual information of time series data
| null | null | null | null |
physics.soc-ph cs.CE cs.IT math.IT stat.AP
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The topology of a power grid is estimated using an information theoretic approach. By modeling the grid as a graph and using voltage magnitude data of individual nodes in the grid, the mutual information between pairs of nodes is computed using different approximation methods. Using the well-known Chow-Liu algorithm, a maximum spanning tree based on mutual information is computed to estimate the power grid topology. This manuscript explores the application of this method to different datasets and explores the domain of applicability. The data quality, precision, time windows, frequency and the method for calculating the mutual information are varied to see the effect on the successful reconstruction of the graph and it's leaf nodes. Success shown for IEEE networks generated with MATPOWER and data generated using GridLAB-D. The algorithm is then cross-validated on IEEE networks.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 20:09:04 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:28:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Speckhard",
"Daniel T.",
""
]
] |
2505.11592
|
Harsh Mehta
|
Harsh Mehta and Suvodip Mukherjee
|
The Prospect from the Upcoming CMB Experiment LiteBIRD to Discover Axion-like Particles Using Milky Way
|
30 pages, 13 figures, 3 Tables, Published in JCAP
| null | null | null |
astro-ph.CO astro-ph.GA astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The existence of axion-like particles (ALPs) can be probed from their signatures in the Cosmic Microwave Background (CMB) due to the photon-ALP resonant conversion over the mass range of ALPs that matches with the effective mass of photons in the plasma in the astrophysical systems. Such a conversion can also occur in the Milky Way halo and disk and can cause a unique spatial and spectral distortion. The signal is highly non-Gaussian and cannot be measured precisely by the usual power-spectrum approach. We devise a new technique to search for this signal from the upcoming full-sky CMB experiment LiteBIRD using its multi-frequency band using a template-based spatial profile of the ALP distortion signal. This technique captures the large-scale non-Gaussian aspects of the ALP distortion signal in terms of a spatial template and makes it possible to search for any non-zero ALP signal. We show that the inference of the ALP coupling using the template-based technique from LiteBIRD can provide constraints on the coupling constant approximately $ g_{a\gamma} < 6.5 \times 10^{-12} \, \mathrm{GeV}^{-1}$ for ALP masses below $10^{-14}$ eV at 95\% confidence interval which is an order of magnitude better than the current bounds from CERN Axion Solar Telescope (CAST) at $g_{a\gamma} < 6.6 \times 10^{-11} \, \mathrm{GeV}^{-1}$, This shows the capability of future multi-band CMB experiment LiteBIRD in opening the discovery space towards physics beyond the standard model.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 18:00:30 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:22:32 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mehta",
"Harsh",
""
],
[
"Mukherjee",
"Suvodip",
""
]
] |
2505.11709
|
Ryan Hoque
|
Ryan Hoque, Peide Huang, David J. Yoon, Mouli Sivapurapu, Jian Zhang
|
EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video
| null | null | null | null |
cs.CV cs.LG cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Imitation learning for manipulation has a well-known data scarcity problem. Unlike natural language and 2D computer vision, there is no Internet-scale corpus of data for dexterous manipulation. One appealing option is egocentric human video, a passively scalable data source. However, existing large-scale datasets such as Ego4D do not have native hand pose annotations and do not focus on object manipulation. To this end, we use Apple Vision Pro to collect EgoDex: the largest and most diverse dataset of dexterous human manipulation to date. EgoDex has 829 hours of egocentric video with paired 3D hand and finger tracking data collected at the time of recording, where multiple calibrated cameras and on-device SLAM can be used to precisely track the pose of every joint of each hand. The dataset covers a wide range of diverse manipulation behaviors with everyday household objects in 194 different tabletop tasks ranging from tying shoelaces to folding laundry. Furthermore, we train and systematically evaluate imitation learning policies for hand trajectory prediction on the dataset, introducing metrics and benchmarks for measuring progress in this increasingly important area. By releasing this large-scale dataset, we hope to push the frontier of robotics, computer vision, and foundation models. EgoDex is publicly available for download at https://github.com/apple/ml-egodex.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 21:34:47 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 23:48:50 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hoque",
"Ryan",
""
],
[
"Huang",
"Peide",
""
],
[
"Yoon",
"David J.",
""
],
[
"Sivapurapu",
"Mouli",
""
],
[
"Zhang",
"Jian",
""
]
] |
2505.12355
|
Ya Shen
|
Ya Shen, Gang Chen, Hui Ma, and Mengjie Zhang
|
GATES: Cost-aware Dynamic Workflow Scheduling via Graph Attention Networks and Evolution Strategy
|
This paper has been accepted by the 34th International Joint Conference on Artificial Intelligence (IJCAI-2025)
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Cost-aware Dynamic Workflow Scheduling (CADWS) is a key challenge in cloud computing, focusing on devising an effective scheduling policy to efficiently schedule dynamically arriving workflow tasks, represented as Directed Acyclic Graphs (DAG), to suitable virtual machines (VMs). Deep reinforcement learning (DRL) has been widely employed for automated scheduling policy design. However, the performance of DRL is heavily influenced by the design of the problem-tailored policy network and is highly sensitive to hyperparameters and the design of reward feedback. Considering the above-mentioned issues, this study proposes a novel DRL method combining Graph Attention Networks-based policy network and Evolution Strategy, referred to as GATES. The contributions of GATES are summarized as follows: (1) GATES can capture the impact of current task scheduling on subsequent tasks by learning the topological relationships between tasks in a DAG. (2) GATES can assess the importance of each VM to the ready task, enabling it to adapt to dynamically changing VM resources. (3) Utilizing Evolution Strategy's robustness, exploratory nature, and tolerance for delayed rewards, GATES achieves stable policy learning in CADWS. Extensive experimental results demonstrate the superiority of the proposed GATES in CADWS, outperforming several state-of-the-art algorithms. The source code is available at: https://github.com/YaShen998/GATES.
|
[
{
"version": "v1",
"created": "Sun, 18 May 2025 10:38:41 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 01:15:11 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 01:43:10 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Shen",
"Ya",
""
],
[
"Chen",
"Gang",
""
],
[
"Ma",
"Hui",
""
],
[
"Zhang",
"Mengjie",
""
]
] |
2505.13458
|
Bhargav Sriram Siddani
|
B. Siddani, Weiqun Zhang, Andrew Nonaka, John Bell, Ishan Srivastava
|
An adaptive, data-driven multiscale approach for dense granular flows
|
40 pages, 10 figures, 2 tables
| null |
10.1016/j.cma.2025.118294
| null |
cond-mat.soft physics.comp-ph physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
The accuracy of coarse-grained continuum models of dense granular flows is limited by the lack of high-fidelity closure models for granular rheology. One approach to addressing this issue, referred to as the hierarchical multiscale method, is to use a high-fidelity fine-grained model to compute the closure terms needed by the coarse-grained model. The difficulty with this approach is that the overall model can become computationally intractable due to the high computational cost of the high-fidelity model. In this work, we describe a multiscale modeling approach for dense granular flows that utilizes neural networks trained using high-fidelity discrete element method (DEM) simulations to approximate the constitutive granular rheology for a continuum incompressible flow model. Our approach leverages an ensemble of neural networks to estimate predictive uncertainty that allows us to determine whether the rheology at a given point is accurately represented by the neural network model. Additional DEM simulations are only performed when needed, minimizing the number of additional DEM simulations required when updating the rheology. This adaptive coupling significantly reduces the overall computational cost of the approach while controlling the error. In addition, the neural networks are customized to learn regularized rheological behavior to ensure well-posedness of the continuum solution. We first validate the approach using two-dimensional steady-state and decelerating inclined flows. We then demonstrate the efficiency of our approach by modeling three-dimensional sub-aerial granular column collapse for varying initial column aspect ratios, where our multiscale method compares well with the computationally expensive computational fluid dynamics (CFD)-DEM simulation.
|
[
{
"version": "v1",
"created": "Wed, 30 Apr 2025 06:46:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Siddani",
"B.",
""
],
[
"Zhang",
"Weiqun",
""
],
[
"Nonaka",
"Andrew",
""
],
[
"Bell",
"John",
""
],
[
"Srivastava",
"Ishan",
""
]
] |
2505.13585
|
Xinzhu Liang
|
Xinzhu Liang, Joseph M. Lukens, Sanjaya Lohani, Brian T. Kirby, Thomas A. Searles, Xin Qiu, Kody J. H. Law
|
Scalable Bayesian Monte Carlo: fast uncertainty estimation beyond deep ensembles
|
38 pages, 30 figures, 23 tables
| null | null | null |
stat.ML cs.LG stat.CO
|
http://creativecommons.org/licenses/by/4.0/
|
This work introduces a new method designed for Bayesian deep learning called scalable Bayesian Monte Carlo (SBMC). The method is comprised of a model and an algorithm. The model interpolates between a point estimator and the posterior. The algorithm is a parallel implementation of sequential Monte Carlo sampler (SMC$_\parallel$) or Markov chain Monte Carlo (MCMC$_\parallel$). We collectively refer to these consistent (asymptotically unbiased) algorithms as Bayesian Monte Carlo (BMC), and any such algorithm can be used in our SBMC method. The utility of the method is demonstrated on practical examples: MNIST, CIFAR, IMDb. A systematic numerical study reveals that for the same wall-clock time as state-of-the-art (SOTA) methods like deep ensembles (DE), SBMC achieves comparable or better accuracy and substantially improved uncertainty quantification (UQ)--in particular, epistemic UQ. This is demonstrated on the downstream task of estimating the confidence in predictions, which can be used for reliability assessment or abstention decisions.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 17:55:32 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:11:57 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Liang",
"Xinzhu",
""
],
[
"Lukens",
"Joseph M.",
""
],
[
"Lohani",
"Sanjaya",
""
],
[
"Kirby",
"Brian T.",
""
],
[
"Searles",
"Thomas A.",
""
],
[
"Qiu",
"Xin",
""
],
[
"Law",
"Kody J. H.",
""
]
] |
2505.13832
|
Jacob Turner
|
Jacob E. Turner, Timothy Dolch, Paul B. Demorest, Ryan S. Lynch, Daniel R. Stinebring, Cody Jessup, Nathaniel Jones, and Christopher Scheithauer
|
Pulsar Cyclic Spectroscopy in the Partial-Deconvolution Regime: Benefits & Limitations
|
Accepted to ApJ
|
2025 ApJ 989 228
|
10.3847/1538-4357/adec73
| null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
We explore possible advantages of cyclic spectroscopy for observations of pulsars in instances where full cyclic deconvolution is not feasible. We compute cyclic merits and full-deconvolution regime boundaries for pulsars observed by NANOGrav and discuss which sources stand to benefit the most from using cyclic spectroscopy when observed with the Green Bank Telescope and DSA-2000 in a given frequency range. We compare data products, namely the wavefield, in both full-deconvolution and partial-deconvolution regimes to demonstrate what can be accomplished with incomplete phase retrieval. Additionally, we show how some phase retrieval can still be achieved in the partial-deconvolution regime and how this allows for additional information in scintillation-based data products, like the dynamic wavefield power, compared to what can be found in traditional dynamic spectra. An examination of dynamic wavefield phase as a function of observing frequency reveals more complete phase retrieval is achieved the closer one gets to the full deconvolution regime, agreeing with the expectations of cyclic merit. While we demonstrate that fragmentary recovery of the secondary wavefield can be accomplished in the partial-deconvolution regime, we advocate for a synergistic approach with phase retrieval methods like the $\theta-\theta$ transform, although we also provide discussion about shortcomings of this strategy. Finally, we use the combination of modest cyclic merit and lack of discernible results for PSR J1903$+$0327 to motivate the creation of an updated "cyclic merit 2.0", which relies on scintillation bandwidth instead of observing bandwidth.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 02:20:19 GMT"
},
{
"version": "v2",
"created": "Mon, 07 Jul 2025 18:02:46 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 19:19:00 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Turner",
"Jacob E.",
""
],
[
"Dolch",
"Timothy",
""
],
[
"Demorest",
"Paul B.",
""
],
[
"Lynch",
"Ryan S.",
""
],
[
"Stinebring",
"Daniel R.",
""
],
[
"Jessup",
"Cody",
""
],
[
"Jones",
"Nathaniel",
""
],
[
"Scheithauer",
"Christopher",
""
]
] |
2505.13950
|
Ziyang Zeng
|
Ziyang Zeng, Dun Zhang, Jiacheng Li, Panxiang Zou, Yuqing Yang
|
An Empirical Study of Position Bias in Modern Information Retrieval
|
EMNLP 2025 Findings
| null | null | null |
cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
This study investigates the position bias in information retrieval, where models tend to overemphasize content at the beginning of passages while neglecting semantically relevant information that appears later. To analyze the extent and impact of position bias, we introduce a new evaluation framework consisting of two position-aware retrieval benchmarks (SQuAD-PosQ, FineWeb-PosQ) and an intuitive diagnostic metric, the Position Sensitivity Index (PSI), for quantifying position bias from a worst-case perspective. We conduct a comprehensive evaluation across the full retrieval pipeline, including BM25, dense embedding models, ColBERT-style late-interaction models, and full-interaction reranker models. Our experiments show that when relevant information appears later in the passage, dense embedding models and ColBERT-style models suffer significant performance degradation (an average drop of 15.6%). In contrast, BM25 and reranker models demonstrate greater robustness to such positional variation. These findings provide practical insights into model sensitivity to the position of relevant information and offer guidance for building more position-robust retrieval systems. Code and data are publicly available at: https://github.com/NovaSearch-Team/position-bias-in-IR.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 05:29:01 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:51:03 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zeng",
"Ziyang",
""
],
[
"Zhang",
"Dun",
""
],
[
"Li",
"Jiacheng",
""
],
[
"Zou",
"Panxiang",
""
],
[
"Yang",
"Yuqing",
""
]
] |
2505.14477
|
Maria Panagiotou
|
Maria Panagiotou, Lorenzo Brigato, Vivien Streit, Amanda Hayoz, Stephan Proennecke, Stavros Athanasopoulos, Mikkel T. Olsen, Elizabeth J. den Brok, Cecilie H. Svensson, Konstantinos Makrilakis, Maria Xatzipsalti, Andriani Vazeou, Peter R. Mertens, Ulrik Pedersen-Bjergaard, Bastiaan E. de Galan, and Stavroula Mougiakakou
|
Personalised Insulin Adjustment with Reinforcement Learning: An In-Silico Validation for People with Diabetes on Intensive Insulin Treatment
| null | null |
10.1109/ACCESS.2025.3600738
|
2025.3600738
|
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Despite recent advances in insulin preparations and technology, adjusting insulin remains an ongoing challenge for the majority of people with type 1 diabetes (T1D) and longstanding type 2 diabetes (T2D). In this study, we propose the Adaptive Basal-Bolus Advisor (ABBA), a personalised insulin treatment recommendation approach based on reinforcement learning for individuals with T1D and T2D, performing self-monitoring blood glucose measurements and multiple daily insulin injection therapy. We developed and evaluated the ability of ABBA to achieve better time-in-range (TIR) for individuals with T1D and T2D, compared to a standard basal-bolus advisor (BBA). The in-silico test was performed using an FDA-accepted population, including 101 simulated adults with T1D and 101 with T2D. An in-silico evaluation shows that ABBA significantly improved TIR and significantly reduced both times below- and above-range, compared to BBA. ABBA's performance continued to improve over two months, whereas BBA exhibited only modest changes. This personalised method for adjusting insulin has the potential to further optimise glycaemic control and support people with T1D and T2D in their daily self-management. Our results warrant ABBA to be trialed for the first time in humans.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 15:10:06 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Panagiotou",
"Maria",
""
],
[
"Brigato",
"Lorenzo",
""
],
[
"Streit",
"Vivien",
""
],
[
"Hayoz",
"Amanda",
""
],
[
"Proennecke",
"Stephan",
""
],
[
"Athanasopoulos",
"Stavros",
""
],
[
"Olsen",
"Mikkel T.",
""
],
[
"Brok",
"Elizabeth J. den",
""
],
[
"Svensson",
"Cecilie H.",
""
],
[
"Makrilakis",
"Konstantinos",
""
],
[
"Xatzipsalti",
"Maria",
""
],
[
"Vazeou",
"Andriani",
""
],
[
"Mertens",
"Peter R.",
""
],
[
"Pedersen-Bjergaard",
"Ulrik",
""
],
[
"de Galan",
"Bastiaan E.",
""
],
[
"Mougiakakou",
"Stavroula",
""
]
] |
2505.14910
|
Yu Zhang
|
Yu Zhang, Wenxiang Guo, Changhao Pan, Dongyu Yao, Zhiyuan Zhu, Ziyue Jiang, Yuhan Wang, Tao Jin, Zhou Zhao
|
TCSinger 2: Customizable Multilingual Zero-shot Singing Voice Synthesis
|
Accepted by Findings of ACL 2025
|
Findings of the Association for Computational Linguistics: ACL 2025
|
10.18653/v1/2025.findings-acl.687
| null |
eess.AS cs.CL cs.SD
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Customizable multilingual zero-shot singing voice synthesis (SVS) has various potential applications in music composition and short video dubbing. However, existing SVS models overly depend on phoneme and note boundary annotations, limiting their robustness in zero-shot scenarios and producing poor transitions between phonemes and notes. Moreover, they also lack effective multi-level style control via diverse prompts. To overcome these challenges, we introduce TCSinger 2, a multi-task multilingual zero-shot SVS model with style transfer and style control based on various prompts. TCSinger 2 mainly includes three key modules: 1) Blurred Boundary Content (BBC) Encoder, predicts duration, extends content embedding, and applies masking to the boundaries to enable smooth transitions. 2) Custom Audio Encoder, uses contrastive learning to extract aligned representations from singing, speech, and textual prompts. 3) Flow-based Custom Transformer, leverages Cus-MOE, with F0 supervision, enhancing both the synthesis quality and style modeling of the generated singing voice. Experimental results show that TCSinger 2 outperforms baseline models in both subjective and objective metrics across multiple related tasks. Singing voice samples are available at https://aaronz345.github.io/TCSinger2Demo/.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 21:04:10 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 00:54:20 GMT"
},
{
"version": "v3",
"created": "Fri, 30 May 2025 10:20:59 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhang",
"Yu",
""
],
[
"Guo",
"Wenxiang",
""
],
[
"Pan",
"Changhao",
""
],
[
"Yao",
"Dongyu",
""
],
[
"Zhu",
"Zhiyuan",
""
],
[
"Jiang",
"Ziyue",
""
],
[
"Wang",
"Yuhan",
""
],
[
"Jin",
"Tao",
""
],
[
"Zhao",
"Zhou",
""
]
] |
2505.14944
|
Rodolfo Maza
|
Rodolfo E. Maza
|
Periodic operators over a component domain and homogenization of some class of quasi-linear elliptic problems in two-component domain with interfacial resistance
|
Added clarifications on the purpose of defining an appropriate extension
| null | null | null |
math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
This paper addresses the periodic homogenization of quasilinear elliptic PDEs in a two-component domain with an interfacial thermal barrier. It introduces a periodic extension operator that ensures strong convergence of function sequences in the Sobolev space. Moreover, two families of quasilinear elliptic problems in two-component domains with interfacial resistance will be considered here. One family with \(L^2\) data and another family with \(L^1\) data.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 22:04:44 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:59:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Maza",
"Rodolfo E.",
""
]
] |
2505.15609
|
Chih-Chun Chien
|
Xin Wang, Hao Guo, and Chih-Chun Chien
|
Uhlmann and scalar Wilczek-Zee phases of degenerate quantum systems
|
17 pages, 2 figures, submitted
| null | null | null |
quant-ph cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Wilczek-Zee (WZ) holonomy arises in degenerate states while the Uhlmann holonomy characterizes finite-temperature topology. We investigate possible relationships between the Uhlmann phase and the scalar WZ phase, which reflects the Uhlmann and WZ holonomy respectively, in an exemplary four-level model with two doubly degenerate subspaces. Through exact solutions, we contrast the behavior of the Uhlmann and WZ connections and their associated phases. In the zero-temperature limit, the Uhlmann phase may or may not agree with the scalar WZ phase of the degenerate ground states due to obstructions from the Hamiltonian manifested as Dirac points. This is in stark contrast to non-degenerate systems where the correspondence between the Uhlmann and Berry phases in general holds. Our analyses further show that for the example studied here, the Uhlmann phase catches the singular behavior at the Dirac points while the WZ connection and scalar WZ phase vanish along a zero-field axis. We also briefly discuss possible experimental implications.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 15:01:52 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:07:00 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Xin",
""
],
[
"Guo",
"Hao",
""
],
[
"Chien",
"Chih-Chun",
""
]
] |
2505.16546
|
Jun Maeda
|
Jun Maeda, Yuya Tanizaki
|
Twisted Partition Functions as Order Parameters
|
33 pages, references added
|
JHEP 08 (2025) 128
|
10.1007/JHEP08(2025)128
|
YITP-25-42, KUNS-3051
|
hep-th cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For quantum field theories with global symmetry, we can study the behavior of the partition function with the background gauge field to diagnose different quantum phases. For the case of discrete symmetries, we find that the symmetry-twisted partition function works as an order parameter that discriminates spontaneous symmetry breaking (SSB), symmetry-protected topological (SPT) states, and symmetry-enriched topological (SET) states. We then consider its application to the case of 4d Yang-Mills theory with adjoint matters to understand the relation between the twisted partition function and the Wilson-'t Hooft classification. We also study its behavior for the spontaneously broken U(1) symmetry and interpret the result from the viewpoint of the mixed anomaly with the emergent solitonic symmetry.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 11:37:23 GMT"
},
{
"version": "v2",
"created": "Sun, 01 Jun 2025 14:51:25 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Maeda",
"Jun",
""
],
[
"Tanizaki",
"Yuya",
""
]
] |
2505.17559
|
Zhufeng Yao
|
Zhufeng Yao
|
Critical Exponent Rigidity for $\Theta-$positive Representations
|
Ver 1: 39 pages Ver 2: 42 pages; Theorem 1.4 in the first version was not true, we have modified it to the correct statement and adjusted the proof in Section 4.3
| null | null | null |
math.DG math.DS math.GR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove for a $\Theta-$positive representation from a discrete subgroup $\Gamma\subset \mathsf{PSL}(2,\mathbb{R})$, the critical exponent for any $\alpha\in \Theta$ is not greater than one. When $\Gamma$ is geometrically finite, the equality holds if and only if $\Gamma$ is a lattice.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 07:05:24 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 03:16:35 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yao",
"Zhufeng",
""
]
] |
2505.17894
|
Sara Chrouf
|
Khalil Hennara, Muhammad Hreden, Mohamed Motaism Hamed, Zeina Aldallal, Sara Chrouf, and Safwan AlModhayan
|
Mutarjim: Advancing Bidirectional Arabic-English Translation with a Small Language Model
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We introduce Mutarjim, a compact yet powerful language model for bidirectional Arabic-English translation. While large-scale LLMs have shown impressive progress in natural language processing tasks, including machine translation, smaller models. Leveraging this insight, we developed Mutarjim based on Kuwain-1.5B , a language model tailored for both Arabic and English. Despite its modest size, Mutarjim outperforms much larger models on several established benchmarks, achieved through an optimized two-phase training approach and a carefully curated, high-quality training corpus.. Experimental results show that Mutarjim rivals models up to 20 times larger while significantly reducing computational costs and training requirements. We also introduce Tarjama-25, a new benchmark designed to overcome limitations in existing Arabic-English benchmarking datasets, such as domain narrowness, short sentence lengths, and English-source bias. Tarjama-25 comprises 5,000 expert-reviewed sentence pairs and spans a wide range of domains, offering a more comprehensive and balanced evaluation framework. Notably, Mutarjim achieves state-of-the-art performance on the English-to-Arabic task in Tarjama-25, surpassing even significantly larger and proprietary models like GPT-4o mini. We publicly release Tarjama-25 to support future research and advance the evaluation of Arabic-English translation systems.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 13:42:21 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:51:52 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hennara",
"Khalil",
""
],
[
"Hreden",
"Muhammad",
""
],
[
"Hamed",
"Mohamed Motaism",
""
],
[
"Aldallal",
"Zeina",
""
],
[
"Chrouf",
"Sara",
""
],
[
"AlModhayan",
"Safwan",
""
]
] |
2505.18205
|
Scott McKinley
|
Richard B. Lehoucq, Scott A. McKinley, Petr Plech\'a\v{c}
|
Source identification via pathwise gradient estimation
| null | null | null | null |
math.OC math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
In the context of PDE-constrained optimization theory, source identification problems traditionally entail particles emerging from an unknown source distribution inside a domain, moving according to a prescribed stochastic process, e.g.~Brownian motion, and then exiting through the boundary of a compact domain. Given information about the flux of particles through the boundary of the domain, the challenge is to infer as much as possible about the source.
In the PDE setting, it is usually assumed that the flux can be observed without error and at all points on the boundary. Here we consider a different, more statistical presentation of the problem, in which the data has the form of discrete counts of particles arriving at a set of disjoint detectors whose union is a strict subset of the boundary. In keeping with the primacy of the stochastic processes in the generation of the model, we present a stochastic gradient descent algorithm in which exit rates and parameter sensitivities are computed by simulations of particle paths. We present examples for both It\^o diffusion and piecewise-deterministic Markov processes, noting that the form of the sensitivities depends only on the parameterization of the source distribution and is universal among a large class of Markov processes.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 16:27:09 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:18:13 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Lehoucq",
"Richard B.",
""
],
[
"McKinley",
"Scott A.",
""
],
[
"Plecháč",
"Petr",
""
]
] |
2505.18937
|
Mark Hertzberg
|
Zahra Bayat, Mark P. Hertzberg
|
Examining Quintessence Models with DESI Data
|
9 pages in double column format. 5 figures. 2 tables. V2: Updated statistics, added references. Accepted for publication in JCAP
|
JCAP 08 (2025) 065
|
10.1088/1475-7516/2025/08/065
| null |
astro-ph.CO gr-qc hep-ph hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We examine data from the Dark Energy Spectroscopic Instrument (DESI) collaboration which has implications for the nature of dark energy. We consider classes of models that manifestly obey the null energy condition, with a focus on quintessence models. We find that hilltop potentials and exponential potentials provide modest improvement compared to a cosmological constant, but the statistical evidence is only marginal at this stage. We correct some analyses in the existing literature which attempted to compare some quintessence models to the data, giving an overly positive result.
|
[
{
"version": "v1",
"created": "Sun, 25 May 2025 02:09:41 GMT"
},
{
"version": "v2",
"created": "Sun, 06 Jul 2025 08:35:15 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Bayat",
"Zahra",
""
],
[
"Hertzberg",
"Mark P.",
""
]
] |
2505.19686
|
Manex Alkorta
|
Manex Alkorta, Martin Gutierrez-Amigo, {\DH}or{\dj}e Dangi\'c, Chunyu Guo, Philip J. W. Moll, Maia G. Vergniory and Ion Errea
|
Symmetry-broken ground state and phonon mediated superconductivity in Kagome CsV$_3$Sb$_5$
|
8 pages, 5 figures
| null | null | null |
cond-mat.mtrl-sci
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The newly discovered family of non-magnetic Kagome metals AV$_3$Sb$_5$ (A=K,Rb,Cs) provides a unique platform for exploring the interplay between charge density wave (CDW) order, superconductivity, non-trivial topology, and spontaneous time-reversal symmetry breaking. Although characterizing the CDW phase is essential for understanding and modeling these exotic phenomena, its nature remains unresolved. In this work, we employ first-principles free-energy calculations, accounting for both ionic kinetic energy and anharmonic effects, to resolve the atomistic phase diagram of CsV$_3$Sb$_5$ and its charge ordering structure. Our results uncover that the CDW ground state is formed by reconstructed vanadium Kagome layers in a triangular hexagonal pattern, featuring energetically degenerate different stacking orders. This accounts for the various out-of-plane modulations observed experimentally and supports the coexistence of multiple domains. The discovered symmetry-broken ground state is consistent with the absence of any electronic anisotropy in transport experiments. By combining anharmonic phonons with the calculation of electron-phonon matrix elements, we predict a superconducting critical temperature for the CDW phase in agreement with experiments, showing that superconductivity is phonon mediated. These findings not only resolve a long-standing structural puzzle, but also clarify the impact of the CDW in superconductivity, highlighting its fundamental importance in shaping the low-temperature quantum phase diagram of Kagome metals.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 08:45:45 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 17:52:37 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 09:08:44 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Alkorta",
"Manex",
""
],
[
"Gutierrez-Amigo",
"Martin",
""
],
[
"Dangić",
"Ðorđe",
""
],
[
"Guo",
"Chunyu",
""
],
[
"Moll",
"Philip J. W.",
""
],
[
"Vergniory",
"Maia G.",
""
],
[
"Errea",
"Ion",
""
]
] |
2505.20282
|
Zitian Gao
|
Zitian Gao, Lynx Chen, Haoming Luo, Joey Zhou, Bryan Dai
|
One-shot Entropy Minimization
|
Work in progress
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We trained 13,440 large language models and found that entropy minimization requires only a single unlabeled data and 10 steps optimization to achieve performance improvements comparable to or even greater than those obtained using thousands of data and carefully designed rewards in rule-based reinforcement learning. This striking result may prompt a rethinking of post-training paradigms for large language models. Our code is avaliable at https://github.com/zitian-gao/one-shot-em.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 17:58:30 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 17:18:18 GMT"
},
{
"version": "v3",
"created": "Tue, 03 Jun 2025 17:45:49 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 08:51:49 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Gao",
"Zitian",
""
],
[
"Chen",
"Lynx",
""
],
[
"Luo",
"Haoming",
""
],
[
"Zhou",
"Joey",
""
],
[
"Dai",
"Bryan",
""
]
] |
2505.21509
|
Carlos M Portela
|
Somayajulu Dhulipala, Carlos M. Portela
|
Curvature-Guided Mechanics and Design of Spinodal and Shell-Based Architected Materials
|
32 pages, 10 figures, supplementary appendix
| null |
10.1016/j.jmps.2025.106273
| null |
cond-mat.mtrl-sci cond-mat.soft physics.app-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Additively manufactured (AM) architected materials have enabled unprecedented control over mechanical properties of engineered materials. While lattice architectures have played a key role in these advances, they suffer from stress concentrations at sharp joints and bending-dominated behavior at high relative densities, limiting their mechanical efficiency. Additionally, high-resolution AM techniques often result in low-throughput or costly fabrication, restricting manufacturing scalability of these materials. Aperiodic spinodal architected materials offer a promising alternative by leveraging low-curvature architectures that can be fabricated through techniques beyond AM. Enabled by phase separation processes, these architectures exhibit tunable mechanical properties and enhanced defect tolerance by tailoring their curvature distributions. However, the relation between curvature and their anisotropic mechanical behavior remains poorly understood. In this work, we develop a theoretical framework to quantify the role of curvature in governing the anisotropic stiffness and strength of shell-based spinodal architected materials. We introduce geometric metrics that predict the distribution of stretching and bending energies under different loading conditions, bridging the gap between curvature in doubly curved shell-based morphologies and their mechanical anisotropy. We validate our framework through finite element simulations and microscale experiments, demonstrating its utility in designing mechanically robust spinodal architectures. This study provides fundamental insights into curvature-driven mechanics, guiding the optimization of next-generation architected materials for engineering applications.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 15:57:22 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Dhulipala",
"Somayajulu",
""
],
[
"Portela",
"Carlos M.",
""
]
] |
2505.22146
|
Guangfu Hao
|
Guangfu Hao, Haojie Wen, Liangxuan Guo, Yang Chen, Yanchao Bi, Shan Yu
|
Flexible Tool Selection through Low-dimensional Attribute Alignment of Vision and Language
| null | null | null | null |
cs.CV cs.AI cs.CL q-bio.NC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Flexible tool selection reflects a complex cognitive ability that distinguishes humans from other species, yet computational models that capture this ability remain underdeveloped. We developed a framework using low-dimensional attribute representations to bridge visual tool perception and linguistic task understanding. We constructed a comprehensive dataset (ToolNet) containing 115 common tools labeled with 13 carefully designed attributes spanning physical, functional, and psychological properties, paired with natural language scenarios describing tool usage. Visual encoders (ResNet or ViT) extract attributes from tool images while fine-tuned language models (GPT-2, LLaMA, DeepSeek) derive required attributes from task descriptions. Our approach achieves 74% accuracy in tool selection tasks-significantly outperforming direct tool matching (20%) and smaller multimodal models (21%-58%), while approaching performance of much larger models like GPT-4o (73%) with substantially fewer parameters. Human evaluation studies validate our framework's alignment with human decision-making patterns, and generalization experiments demonstrate effective performance on novel tool categories. Ablation studies revealed that manipulation-related attributes (graspability, elongation, hand-relatedness) consistently prove most critical across modalities. This work provides a parameter-efficient, interpretable solution that mimics human-like tool cognition, advancing both cognitive science understanding and practical applications in tool selection tasks.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 09:06:04 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Jun 2025 05:15:58 GMT"
},
{
"version": "v3",
"created": "Mon, 18 Aug 2025 08:43:01 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 04:52:25 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hao",
"Guangfu",
""
],
[
"Wen",
"Haojie",
""
],
[
"Guo",
"Liangxuan",
""
],
[
"Chen",
"Yang",
""
],
[
"Bi",
"Yanchao",
""
],
[
"Yu",
"Shan",
""
]
] |
2505.22306
|
Liyuan Wang
|
Zehua Chen, Yuyang Miao, Liyuan Wang, Luyun Fan, Danilo P. Mandic, Jun Zhu
|
Versatile Cardiovascular Signal Generation with a Unified Diffusion Transformer
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Cardiovascular signals such as photoplethysmography (PPG), electrocardiography (ECG), and blood pressure (BP) are inherently correlated and complementary, together reflecting the health of cardiovascular system. However, their joint utilization in real-time monitoring is severely limited by diverse acquisition challenges from noisy wearable recordings to burdened invasive procedures. Here we propose UniCardio, a multi-modal diffusion transformer that reconstructs low-quality signals and synthesizes unrecorded signals in a unified generative framework. Its key innovations include a specialized model architecture to manage the signal modalities involved in generation tasks and a continual learning paradigm to incorporate varying modality combinations. By exploiting the complementary nature of cardiovascular signals, UniCardio clearly outperforms recent task-specific baselines in signal denoising, imputation, and translation. The generated signals match the performance of ground-truth signals in detecting abnormal health conditions and estimating vital signs, even in unseen domains, while ensuring interpretability for human experts. These advantages position UniCardio as a promising avenue for advancing AI-assisted healthcare.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 12:45:39 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 03:53:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Chen",
"Zehua",
""
],
[
"Miao",
"Yuyang",
""
],
[
"Wang",
"Liyuan",
""
],
[
"Fan",
"Luyun",
""
],
[
"Mandic",
"Danilo P.",
""
],
[
"Zhu",
"Jun",
""
]
] |
2505.22315
|
Adam B. Cahaya
|
Adam B. Cahaya
|
Spin current compensation from competing magnon modes in ferrimagnets
| null |
J. Phys. D: Appl. Phys. 58 (2025) 345002
|
10.1088/1361-6463/adfaa0
| null |
cond-mat.mes-hall quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate thermal spin pumping in gadolinium iron garnet (GdIG), focusing on the mode-resolved dynamics of antiferromagnetic magnons and their impact on spin and heat transport. Antiferromagnets support both right-handed and left-handed magnon modes, which we treat as positive and negative frequency branches, analogous to electrons and holes in semiconductors. Using a two-sublattice model with a minimal exchange interaction scheme, we derive the temperature-dependent spin-wave spectrum and evaluate the associated thermal spin pumping coefficients. Our analysis reveals that the competition between left- and right-handed modes gives rise to a compensation temperature, where the net thermally generated spin current vanishes. Importantly, we show that this compensation point does not necessarily coincide with the crossing of magnon dispersion branches. While previous research considers a detailed microscopic model including all magnetic sublattices and exchange couplings, our approach demonstrates that key features of mode-resolved spin transport can be captured by a simplified and analytically transparent model. These findings advance the understanding of spin-caloritronic phenomena in ferrimagnets and offer new perspectives for the design of chiral magnon-based spintronic devices.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 12:59:37 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Jul 2025 21:49:50 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Cahaya",
"Adam B.",
""
]
] |
2505.22712
|
Andrew Marszewski
|
Andrew Marszewski, Claude-Andr\'e Faucher-Gigu\`ere, Robert Feldmann, Guochao Sun
|
Explaining the Weak Evolution of the High-Redshift Mass-Metallicity Relation with Galaxy Burst Cycles
|
18 pages, 9 figures
| null |
10.3847/2041-8213/adf74b
| null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent observations suggest a nearly constant gas-phase mass-metallicity relation (MZR) at $z \gtrsim 5$, in agreement with many theoretical predictions. This lack of evolution contrasts with observations at $z \lesssim 3$, which find an increasing normalization of the MZR with decreasing redshift. We analyze a high-redshift suite of FIRE-2 cosmological zoom-in simulations to identify the physical drivers of the MZR. Previous studies have explained the weak evolution of the high-redshift MZR in terms of weakly evolving or saturated gas fractions, but we find this alone does not explain the evolution in FIRE-2. Instead, stellar feedback following intense bursts of star formation drives enriched gas out of galaxies, resetting their interstellar medium and separating their histories into distinct ``burst cycles". We develop the ``Reduced Burst Model", a simplified gas-regulator model that successfully reproduces the simulated MZR and identifies the dominant drivers of its evolution. As redshift decreases, the metallicity of inflows within burst cycles increases at fixed stellar mass due to increased wind recycling of enriched gas. Meanwhile, the metal mass produced by stars per inflowing gas mass within these cycles decreases because of decreased star formation per gas mass inflowing into the galaxy. The effects of these two processes on the median metallicity largely cancel, holding the MZR constant for $z = 5 - 12$. At fixed stellar mass, the simulations predict lower gas metallicities at higher $\rm H\alpha$-derived star formation rates, in qualitative agreement with the fundamental metallicity relation (FMR), but this effect is reduced in rest UV-selected samples.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 18:04:59 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Marszewski",
"Andrew",
""
],
[
"Faucher-Giguère",
"Claude-André",
""
],
[
"Feldmann",
"Robert",
""
],
[
"Sun",
"Guochao",
""
]
] |
2505.24349
|
Alisa Krokhmal
|
G. Sakharova, A. Krokhmal, R. Galimova, A. Khatmullina, D. Nabiullina, I. Buzaev, D. Avzaletdinova, D. Chupova, V. Khokhlova
|
The Use of Alendronate to Enhance Transcranial Transmission of Focused Ultrasound for Successful Ablations in Brain
| null | null |
10.1016/j.ultras.2025.107796
| null |
physics.med-ph physics.data-an
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Objective: The aim of this study was to evaluate the efficacy of alendronate therapy in improving bone density distribution in skull bones and corresponding ultrasound permeability in patients who had previously experienced unsuccessful transcranial MR-guided focused ultrasound (MRgFUS) ablation. The ability of alendronate treatment to modify skull bone characteristics and enhance the success rate of repeat MRgFUS procedures was assessed.
Methods: Five patients with initially unsuccessful MRgFUS ablations underwent a 6-12 month regimen of alendronate to improve bone density. Repeat MRgFUS procedures were performed, and changes in skull density ratio (SDR) and peak focal temperatures were evaluated statistically using CT and MR imaging. Histograms of skull bone density were introduced and analysed as an additional metric.
Results: After therapy, SDR increased in four out of five patients (from 0.378$\pm$0.037 to 0.424$\pm$0.045, p>0.05). All repeated procedures were successful. The maximum focal temperature, averaged over sonications, increased from 53.6$\pm$4.0{\deg}C to 55.7$\pm$4.1{\deg}C (p=0.018), while the maximum temperature per patient rose from 57.0$\pm$2.4{\deg}C to 60.2$\pm$1.8{\deg}C (p=0.031). Histograms of CT scans showed a reduction in low-density voxels, indicating trabecular bone densification. 3D CT scan registration revealed local density changes, defect filling, and void reduction.
Conclusions: Alendronate therapy enhanced skull bone density distribution and thus ultrasound permeability, which has facilitated successful repeat MRgFUS. By visually analysing CT changes, healthcare professionals can better inform their decision-making regarding repeat surgeries. This method broadens the pool of patients with low SDR eligible for MRgFUS treatment and underscores the potential benefits of alendronate in improving treatment outcomes.
|
[
{
"version": "v1",
"created": "Fri, 30 May 2025 08:43:48 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Sakharova",
"G.",
""
],
[
"Krokhmal",
"A.",
""
],
[
"Galimova",
"R.",
""
],
[
"Khatmullina",
"A.",
""
],
[
"Nabiullina",
"D.",
""
],
[
"Buzaev",
"I.",
""
],
[
"Avzaletdinova",
"D.",
""
],
[
"Chupova",
"D.",
""
],
[
"Khokhlova",
"V.",
""
]
] |
2506.00307
|
John Harvill
|
John Harvill, Ziwei Fan, Hao Wang, Luke Huan, Anoop Deoras, Yizhou Sun, Hao Ding
|
Lossless Token Sequence Compression via Meta-Tokens
|
16 pages, 8 figures
| null | null | null |
cs.CL cs.AI cs.LG
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Existing work on prompt compression for Large Language Models (LLM) focuses on lossy methods that try to maximize the retention of semantic information that is relevant to downstream tasks while significantly reducing the sequence length. In this paper, we introduce a task-agnostic lossless compression technique similar to LZ77 that makes it possible to reduce the input token sequence length on average by 27\% and 18\% for the two evaluation tasks explored here. Given that we use transformer-based LLMs, this equates to 47\% and 33\% less encoding computation, respectively, due to the quadratic nature of attention. The token sequence transformation is trivial to reverse and highlights that no semantic information is lost in the process. We evaluate our proposed approach on two tasks that require strict preservation of semantics/syntax and demonstrate that existing lossy compression methods perform poorly in this setting. We find that our lossless compression technique produces only a small gap in performance compared to using the uncompressed input and posit that larger models and an expanded computing budget would likely erase the gap entirely.
|
[
{
"version": "v1",
"created": "Fri, 30 May 2025 23:32:57 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 23:19:57 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Harvill",
"John",
""
],
[
"Fan",
"Ziwei",
""
],
[
"Wang",
"Hao",
""
],
[
"Huan",
"Luke",
""
],
[
"Deoras",
"Anoop",
""
],
[
"Sun",
"Yizhou",
""
],
[
"Ding",
"Hao",
""
]
] |
2506.00442
|
Yongming Luo
|
Yongming Luo
|
Critical scattering for the nonlinear Schr\"odinger equation on waveguide manifolds
| null | null | null | null |
math.AP math.CA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the small data scattering problem in critical spaces for the nonlinear Schr\"odinger equation (NLS) on waveguide manifolds. Our work is primarily inspired by the recent paper of Kwak and Kwon \cite{KwakKwon} that established the local well-posedness of the periodic NLS with possibly non-algebraic nonlinearity. While we adopt a framework similar to \cite{KwakKwon} for our problem, two main obstacles prevent its direct adaptation to the waveguide setting. First, the classical Strichartz estimates for NLS in critical product spaces, introduced by Hani and Pausader, possess limited endpoints and are thus inapplicable to high-dimensional waveguides. Second, the crucial fractional arguments used in \cite{KwakKwon} rely on a well-known fractional derivative formula due to Strichartz, which admits only a Hilbert space-valued extension and is therefore incompatible with our model setting.
To overcome these difficulties, we develop an anisotropic generalization of the framework in \cite{KwakKwon} using the anisotropic Strichartz estimates established by Tzvetkov and Visciglia, which allow for nearly unlimited endpoints. We also resolve several new challenges arising from the vector-valued and anisotropic nature of the model by employing novel interpolation techniques within Besov spaces. As a further novelty, we provide a new proof of the main result based on classical fixed point arguments, differing from the approximation methods used in \cite{KwakKwon}. Consequently, we settle the small data scattering problem in critical spaces for the NLS with arbitrary mass-supercritical nonlinearity on waveguide manifolds.
|
[
{
"version": "v1",
"created": "Sat, 31 May 2025 07:49:29 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 03:35:21 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Luo",
"Yongming",
""
]
] |
2506.00853
|
Dena Mujtaba
|
Dena Mujtaba, Nihar Mahapatra
|
Fine-Tuning ASR for Stuttered Speech: Personalized vs. Generalized Approaches
|
Accepted to Interspeech 2025
| null |
10.21437/Interspeech.2025-2373
| null |
cs.SD eess.AS
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Stuttering -- characterized by involuntary disfluencies such as blocks, prolongations, and repetitions -- is often misinterpreted by automatic speech recognition (ASR) systems, resulting in elevated word error rates and making voice-driven technologies inaccessible to people who stutter. The variability of disfluencies across speakers and contexts further complicates ASR training, compounded by limited annotated stuttered speech data. In this paper, we investigate fine-tuning ASRs for stuttered speech, comparing generalized models (trained across multiple speakers) to personalized models tailored to individual speech characteristics. Using a diverse range of voice-AI scenarios, including virtual assistants and video interviews, we evaluate how personalization affects transcription accuracy. Our findings show that personalized ASRs significantly reduce word error rates, especially in spontaneous speech, highlighting the potential of tailored models for more inclusive voice technologies.
|
[
{
"version": "v1",
"created": "Sun, 01 Jun 2025 06:25:20 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:59:09 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mujtaba",
"Dena",
""
],
[
"Mahapatra",
"Nihar",
""
]
] |
2506.02251
|
Christopher S. Hill
|
S. Alcott, Z. Bhatti, J. Brooke, C. Campagnari, M. Carrigan, M. Citron, R. De Los Santos, A. De Roeck, C. Dorofeev, T. Du, J. Goldstein, F. Golf, N. Gonzalez, A. Haas, J. Heymann, C.S. Hill, D. Imani, M. Joyce, K. Larina, R. Loos, S. Lowette, H. Mei, D.W. Miller, B. Peng, S.N. Santpu, I. Reed, E. Schaffer, R. Schmitz, J. Steenis, D. Stuart, J.S. Tafoya Vargas, D. Vannerom, T. Wybouw, Z. Xiao, H. Zaraket, G. Zecchinelli, C. Zheng
|
Search for millicharged particles in proton-proton collisions at $\sqrt{s} = 13.6$ TeV
| null | null |
10.1103/rtp1-679h
|
CERN-EP-2025-120
|
hep-ex
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We report on a search for elementary particles with charges much smaller than the electron charge using a data sample of proton-proton collisions provided by the CERN Large Hadron Collider in 2023--24, corresponding to an integrated luminosity of 124.7~fb$^{-1}$ at a center-of-mass energy of 13.6~TeV. The analysis presented uses the completed Run 3 milliQan bar detector to set the most stringent constraints to date for particles with charges $\leq0.24~\rm{e}$ and masses $\geq0.45~\rm{GeV}$.
|
[
{
"version": "v1",
"created": "Mon, 02 Jun 2025 20:55:15 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:05:14 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Alcott",
"S.",
""
],
[
"Bhatti",
"Z.",
""
],
[
"Brooke",
"J.",
""
],
[
"Campagnari",
"C.",
""
],
[
"Carrigan",
"M.",
""
],
[
"Citron",
"M.",
""
],
[
"Santos",
"R. De Los",
""
],
[
"De Roeck",
"A.",
""
],
[
"Dorofeev",
"C.",
""
],
[
"Du",
"T.",
""
],
[
"Goldstein",
"J.",
""
],
[
"Golf",
"F.",
""
],
[
"Gonzalez",
"N.",
""
],
[
"Haas",
"A.",
""
],
[
"Heymann",
"J.",
""
],
[
"Hill",
"C. S.",
""
],
[
"Imani",
"D.",
""
],
[
"Joyce",
"M.",
""
],
[
"Larina",
"K.",
""
],
[
"Loos",
"R.",
""
],
[
"Lowette",
"S.",
""
],
[
"Mei",
"H.",
""
],
[
"Miller",
"D. W.",
""
],
[
"Peng",
"B.",
""
],
[
"Santpu",
"S. N.",
""
],
[
"Reed",
"I.",
""
],
[
"Schaffer",
"E.",
""
],
[
"Schmitz",
"R.",
""
],
[
"Steenis",
"J.",
""
],
[
"Stuart",
"D.",
""
],
[
"Vargas",
"J. S. Tafoya",
""
],
[
"Vannerom",
"D.",
""
],
[
"Wybouw",
"T.",
""
],
[
"Xiao",
"Z.",
""
],
[
"Zaraket",
"H.",
""
],
[
"Zecchinelli",
"G.",
""
],
[
"Zheng",
"C.",
""
]
] |
2506.02873
|
Matthew Kowal
|
Matthew Kowal, Jasper Timm, Jean-Francois Godbout, Thomas Costello, Antonio A. Arechar, Gordon Pennycook, David Rand, Adam Gleave, Kellin Pelrine
|
It's the Thought that Counts: Evaluating the Attempts of Frontier LLMs to Persuade on Harmful Topics
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Persuasion is a powerful capability of large language models (LLMs) that both enables beneficial applications (e.g. helping people quit smoking) and raises significant risks (e.g. large-scale, targeted political manipulation). Prior work has found models possess a significant and growing persuasive capability, measured by belief changes in simulated or real users. However, these benchmarks overlook a crucial risk factor: the propensity of a model to attempt to persuade in harmful contexts. Understanding whether a model will blindly ``follow orders'' to persuade on harmful topics (e.g. glorifying joining a terrorist group) is key to understanding the efficacy of safety guardrails. Moreover, understanding if and when a model will engage in persuasive behavior in pursuit of some goal is essential to understanding the risks from agentic AI systems. We propose the Attempt to Persuade Eval (APE) benchmark, that shifts the focus from persuasion success to persuasion attempts, operationalized as a model's willingness to generate content aimed at shaping beliefs or behavior. Our evaluation framework probes frontier LLMs using a multi-turn conversational setup between simulated persuader and persuadee agents. APE explores a diverse spectrum of topics including conspiracies, controversial issues, and non-controversially harmful content. We introduce an automated evaluator model to identify willingness to persuade and measure the frequency and context of persuasive attempts. We find that many open and closed-weight models are frequently willing to attempt persuasion on harmful topics and that jailbreaking can increase willingness to engage in such behavior. Our results highlight gaps in current safety guardrails and underscore the importance of evaluating willingness to persuade as a key dimension of LLM risk. APE is available at github.com/AlignmentResearch/AttemptPersuadeEval
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 13:37:51 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 17:50:56 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 22:30:52 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Kowal",
"Matthew",
""
],
[
"Timm",
"Jasper",
""
],
[
"Godbout",
"Jean-Francois",
""
],
[
"Costello",
"Thomas",
""
],
[
"Arechar",
"Antonio A.",
""
],
[
"Pennycook",
"Gordon",
""
],
[
"Rand",
"David",
""
],
[
"Gleave",
"Adam",
""
],
[
"Pelrine",
"Kellin",
""
]
] |
2506.03159
|
Lesley Wheat
|
Lesley Wheat, Martin v. Mohrenschildt and Saeid Habibi
|
Bayes Error Rate Estimation in Difficult Situations
|
23 pages, 13 figures, 20 tables
| null | null | null |
cs.LG stat.ME stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Bayes Error Rate (BER) is the fundamental limit on the achievable generalizable classification accuracy of any machine learning model due to inherent uncertainty within the data. BER estimators offer insight into the difficulty of any classification problem and set expectations for optimal classification performance. In order to be useful, the estimators must also be accurate with a limited number of samples on multivariate problems with unknown class distributions. To determine which estimators meet the minimum requirements for "usefulness", an in-depth examination of their accuracy is conducted using Monte Carlo simulations with synthetic data in order to obtain their confidence bounds for binary classification. To examine the usability of the estimators for real-world applications, new non-linear multi-modal test scenarios are introduced. In each scenario, 2500 Monte Carlo simulations per scenario are run over a wide range of BER values. In a comparison of k-Nearest Neighbor (kNN), Generalized Henze-Penrose (GHP) divergence and Kernel Density Estimation (KDE) techniques, results show that kNN is overwhelmingly the more accurate non-parametric estimator. In order to reach the target of an under 5% range for the 95% confidence bounds, the minimum number of required samples per class is 1000. As more features are added, more samples are needed, so that 2500 samples per class are required at only 4 features. Other estimators do become more accurate than kNN as more features are added, but continuously fail to meet the target range.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 19:58:20 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:19:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wheat",
"Lesley",
""
],
[
"Mohrenschildt",
"Martin v.",
""
],
[
"Habibi",
"Saeid",
""
]
] |
2506.03302
|
James Bagrow
|
James Bagrow and Josh Bongard
|
Multi-Exit Kolmogorov-Arnold Networks: enhancing accuracy and parsimony
|
15 pages, 6 figures, 2 tables
|
Mach. Learn.: Sci. Technol. 6 035037 (2025)
|
10.1088/2632-2153/adf9bd
| null |
cs.LG cs.NE physics.data-an stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Kolmogorov-Arnold Networks (KANs) uniquely combine high accuracy with interpretability, making them valuable for scientific modeling. However, it is unclear a priori how deep a network needs to be for any given task, and deeper KANs can be difficult to optimize and interpret. Here we introduce multi-exit KANs, where each layer includes its own prediction branch, enabling the network to make accurate predictions at multiple depths simultaneously. This architecture provides deep supervision that improves training while discovering the right level of model complexity for each task. Multi-exit KANs consistently outperform standard, single-exit versions on synthetic functions, dynamical systems, and real-world datasets. Remarkably, the best predictions often come from earlier, simpler exits, revealing that these networks naturally identify smaller, more parsimonious and interpretable models without sacrificing accuracy. To automate this discovery, we develop a differentiable "learning-to-exit" algorithm that balances contributions from exits during training. Our approach offers scientists a practical way to achieve both high performance and interpretability, addressing a fundamental challenge in machine learning for scientific discovery.
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 18:41:30 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:58:06 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Bagrow",
"James",
""
],
[
"Bongard",
"Josh",
""
]
] |
2506.03304
|
Matthew Hennefarth
|
Matthew R. Hennefarth, Younghwan Kim, Bhavnesh Jangid, Jacob Wardzala, Matthew R. Hermes, Donald G. Truhlar and Laura Gagliardi
|
MC-PDFT Nuclear Gradients and L-PDFT Energies with Meta and Hybrid Meta On-Top Functionals for Ground- and Excited-State Geometry Optimization and Vertical Excitation Energies
|
31 pages, 9 figures
| null |
10.1021/acs.jctc.5c00899
| null |
physics.chem-ph quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Multiconfiguration pair-density functional theory (MC-PDFT) is a post-MCSCF multireference electronic-structure method that explicitly models strong electron correlation, and linearized pair-density functional theory (L-PDFT) is a recently developed multi-state extension that can accurately model conical intersections and locally-avoided crossings. Because MC-PDFT and L-PDFT rely on an on-top energy functional, their accuracy depends on the quality of the on-top functional used. Recent work has introduced translated meta-gradient-approximation (meta-GA) on-top functionals, and specifically the MC23 hybrid meta-GA on-top functional, which is the first on-top functional specifically optimized for MC-PDFT. Here we report the derivation and implementation of analytic nuclear gradients for MC-PDFT calculations using meta-GA and hybrid meta-GA on-top functionals. This development also enables analytic nuclear gradients for the widely successful tPBE0 hybrid on-top functional. Because MC-PDFT nuclear-gradient calculations involve the derivative of the on-top functional, this development also enables the use of meta-GA on-top functionals in L-PDFT single-point energy calculations. We use the new capabilities to test MC23 for ground-state geometries, excited-state geometries, and vertical excitation energies of s-trans-butadiene and benzophenone as well as to test MC23, another hybrid meta-GA, and seven other meta-GA on-top functionals for 441 vertical excitation energies. We find MC23 performs the best of all nine meta and hybrid meta functionals for vertical excitation energies and is comparable in accuracy to tPBE0 and to the NEVPT2 multireference wave function method. Additionally, we directly compare our MC-PDFT vertical excitation results to previously computed TD-DFT values and find that MC-PDFT outperforms even the best performing Kohn-Sham density functional.
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 18:44:07 GMT"
},
{
"version": "v2",
"created": "Fri, 18 Jul 2025 03:26:45 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 02:43:50 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hennefarth",
"Matthew R.",
""
],
[
"Kim",
"Younghwan",
""
],
[
"Jangid",
"Bhavnesh",
""
],
[
"Wardzala",
"Jacob",
""
],
[
"Hermes",
"Matthew R.",
""
],
[
"Truhlar",
"Donald G.",
""
],
[
"Gagliardi",
"Laura",
""
]
] |
2506.03818
|
Zhe Wang
|
Dong-Xu Yu, Ke-Qi Zeng, Zhe Wang
|
Understanding Flow Behaviors of Supercooled Liquids by Embodying Solid-Liquid Duality at Particle Level
|
32 pages, 31 figures
| null | null | null |
cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding the flow behaviors of supercooled liquids presents a major challenge in liquid-state physics due to the strong nonlinearity and rich phenomena. To unravel this complexity, we introduce the concept of local configurational relaxation time $\tau_\rm{LC}$, which allows us to embody the solid-liquid duality, proposed by Maxwell for phenomenologically describing materials' response to external load, at the particle level. The spatial distribution of $\tau_\rm{LC}$ in flow is heterogeneous. Depending on the comparison between the local mobility measured by $\tau_\rm{LC}$ and the external shear rate, the shear response of local regions is either solid-like or liquid-like. In this way, $\tau_\rm{LC}$ plays a role similar to the Maxwell time. By applying this microscopic solid-liquid duality to different conditions of shear flow with a wide range of shear rates, we describe the emergence of shear thinning in steady shear, and predict the major characteristics of the transient response to start-up shear. Furthermore, we reveal a clear structural foundation for $\tau_\rm{LC}$ and the solid-liquid duality associated with it by introducing an order parameter extracted from local configuration. Thus, we establish a framework that connects microscopic structure, dynamics, local mechanical response, and flow behaviors for supercooled liquids. Finally, we rationalize our framework in terms of activations from energy basins that are facilitated by shear. This model illustrates how local structure, convection and thermal activation collectively determine $\tau_\rm{LC}$. Notably, it predicts two distinct response groups, which well correspond to the microscopic solid-liquid duality.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 10:41:43 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 04:02:50 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yu",
"Dong-Xu",
""
],
[
"Zeng",
"Ke-Qi",
""
],
[
"Wang",
"Zhe",
""
]
] |
2506.04562
|
Xingpeng Sun
|
Xingpeng Sun, Shiyang Jia, Zherong Pan, Kui Wu, Aniket Bera
|
Handle-based Mesh Deformation Guided By Vision Language Model
|
19 pages
| null | null | null |
cs.GR cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Mesh deformation is a fundamental tool in 3D content manipulation. Despite extensive prior research, existing approaches often suffer from low output quality, require significant manual tuning, or depend on data-intensive training. To address these limitations, we introduce a training-free, handle-based mesh deformation method. % Our core idea is to leverage a Vision-Language Model (VLM) to interpret and manipulate a handle-based interface through prompt engineering. We begin by applying cone singularity detection to identify a sparse set of potential handles. The VLM is then prompted to select both the deformable sub-parts of the mesh and the handles that best align with user instructions. Subsequently, we query the desired deformed positions of the selected handles in screen space. To reduce uncertainty inherent in VLM predictions, we aggregate the results from multiple camera views using a novel multi-view voting scheme. % Across a suite of benchmarks, our method produces deformations that align more closely with user intent, as measured by CLIP and GPTEval3D scores, while introducing low distortion -- quantified via membrane energy. In summary, our approach is training-free, highly automated, and consistently delivers high-quality mesh deformations.
|
[
{
"version": "v1",
"created": "Thu, 05 Jun 2025 02:29:42 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 18:28:36 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Sun",
"Xingpeng",
""
],
[
"Jia",
"Shiyang",
""
],
[
"Pan",
"Zherong",
""
],
[
"Wu",
"Kui",
""
],
[
"Bera",
"Aniket",
""
]
] |
2506.05182
|
Shivani Jayantkumar Upadhyay
|
Shivani Upadhyay, Messiah Ataey, Syed Shariyar Murtaza, Yifan Nie, Jimmy Lin
|
On the Comprehensibility of Multi-structured Financial Documents using LLMs and Pre-processing Tools
|
15 pages, 5 figures, 9 tables
| null | null | null |
cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The proliferation of complex structured data in hybrid sources, such as PDF documents and web pages, presents unique challenges for current Large Language Models (LLMs) and Multi-modal Large Language Models (MLLMs) in providing accurate answers. Despite the recent advancements of MLLMs, they still often falter when interpreting intricately structured information, such as nested tables and multi-dimensional plots, leading to hallucinations and erroneous outputs. This paper explores the capabilities of LLMs and MLLMs in understanding and answering questions from complex data structures found in PDF documents by leveraging industrial and open-source tools as part of a pre-processing pipeline. Our findings indicate that GPT-4o, a popular MLLM, achieves an accuracy of 56% on multi-structured documents when fed documents directly, and that integrating pre-processing tools raises the accuracy of LLMs to 61.3% for GPT-4o and 76% for GPT-4, and with lower overall cost. The code is publicly available at https://github.com/OGCDS/FinancialQA.
|
[
{
"version": "v1",
"created": "Thu, 05 Jun 2025 15:52:44 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:52:35 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Upadhyay",
"Shivani",
""
],
[
"Ataey",
"Messiah",
""
],
[
"Murtaza",
"Syed Shariyar",
""
],
[
"Nie",
"Yifan",
""
],
[
"Lin",
"Jimmy",
""
]
] |
2506.05669
|
Roni Muslim
|
Roni Muslim, Jihye Kim, Noriko Oikawa, Rinto Anugraha N Q Z, Zulkaida Akbar
|
Ordering-disordering dynamics of the $q$-voter model under random external bias
|
19 pages, 14 figures
| null | null | null |
physics.soc-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate a variant of the two-state $q$-voter model in which agents update their states under a random external field (which points upward with probability $s$ and downward with probability $1-s$) with probability $p$ or adopt the unanimous opinion of $q$ randomly selected neighbors with probability $ 1-p$. Using mean-field analysis and Monte Carlo simulations, we identify an order-disorder transition at $p_c$ when $s=\tfrac{1}{2}$. Notably, in the regime of $p>p_c$, we estimate the time for systems to reach disordered state from consensus state and find the logarithmic scaling $T_{\text{dis}} \sim \mathcal{B}\ln N$, with $\mathcal{B} = 1/(2p)$ for $q = 1$, while for $q > 1$, $\mathcal{B}$ depends on both $p > p_c$ and $q$. We observe that disordering dynamics slow down significantly for nonlinear strengths $q$ between $2$ and $3$, independent of the probability $p$. On the other hand, when $s=0$ or $s=1$, the system is bound to reach consensus, with the consensus time scaling logarithmically with system size as $T_{\text{con}} \sim \mathcal{B}\ln N$, where $\mathcal{B} = 1/p$ for $q = 1$ and $\mathcal{B} = 1$ for $q > 1$. Furthermore, in the limit of $p = 0$, we derive a closed-form exit probability valid for arbitrary values of $q$ and demonstrate a finite-size scaling collapse. These results clarify how external cues and peer conformity jointly control ordering and disordering in binary opinion dynamics.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 01:32:52 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:56:06 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Muslim",
"Roni",
""
],
[
"Kim",
"Jihye",
""
],
[
"Oikawa",
"Noriko",
""
],
[
"Z",
"Rinto Anugraha N Q",
""
],
[
"Akbar",
"Zulkaida",
""
]
] |
2506.06382
|
Michal Karpowicz Dr
|
Micha{\l} P. Karpowicz
|
On the Fundamental Impossibility of Hallucination Control in Large Language Models
|
cleared mathematics, proofs debugged and ideas expanded, added missing definitions and axioms, discussion and speculation section reorganized, related work improved
| null | null | null |
stat.ML cs.AI cs.CL cs.GT cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
This paper establishes a fundamental impossibility theorem: no LLM capable of performing non-trivial knowledge aggregation can simultaneously achieve truthful knowledge representation, semantic information conservation, complete revelation of relevant knowledge, and knowledge-constrained optimality. The impossibility is not an engineering limitation but arises from the mathematical structure of information aggregation itself.
We establish this result by describing the inference process as an auction of ideas, where distributed components compete exploiting their partial knowledge to shape responses. The proof spans three independent mathematical domains: mechanism design theory (Green-Laffont), the theory of proper scoring rules (Savage), and direct architectural analysis of transformers (Log-Sum-Exp convexity). In particular, we show how to quantify the creation of overconfident or intuitive responses-the signature of both hallucination and creativity, or imagination.
To support this analysis, we introduce the complementary concepts of the semantic information measure and the emergence operator to model bounded reasoning in a general setting. We prove that while bounded reasoning generates accessible information, providing valuable insights and inspirations, the idealized unconstrained reasoning strictly preserves semantic content.
By demonstrating that hallucination and imagination are mathematically identical phenomena-grounded in departures from truthfulness, semantic information conservation, revelation of relevant knowledge, and knowledge-constrained optimality-we offer a principled foundation for managing these behaviors in advanced AI systems. Finally, we present some speculative ideas to inspire evaluation and refinements of the proposed theory.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 23:28:39 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 12:24:10 GMT"
},
{
"version": "v3",
"created": "Tue, 08 Jul 2025 11:43:16 GMT"
},
{
"version": "v4",
"created": "Wed, 06 Aug 2025 11:34:54 GMT"
},
{
"version": "v5",
"created": "Thu, 21 Aug 2025 08:58:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Karpowicz",
"Michał P.",
""
]
] |
2506.06514
|
Filippo Utro
|
Viacheslav Dubovitskii, Aritra Bose, Filippo Utro, Laxmi Parida
|
On Quantum Random Walks in Biomolecular Networks
| null | null | null | null |
quant-ph q-bio.MN q-bio.QM
|
http://creativecommons.org/licenses/by/4.0/
|
Biomolecular networks, such as protein-protein interactions, gene-gene associations, and cell-cell interactions, offer valuable insights into the complex organization of biological systems. These networks are key to understanding cellular functions, disease mechanisms, and identifying therapeutic targets. However, their analysis is challenged by the high dimensionality, heterogeneity, and sparsity of multi-omics data. Random walk algorithms are widely used to propagate information through disease modules, helping to identify disease-associated genes and uncover relevant biological pathways. In this work, we investigate the limitations of classical random walks and explore the potential of quantum random walks (QRWs) for biomolecular network analysis. We evaluate QRWs in two network-based applications. First, in a gene-gene interaction network associated with asthma, autism, and schizophrenia, QRWs more accurately rank disease-associated genes compared to classical methods. Second, in a structured multi-partite cell-cell interaction network derived from mouse brown adipose tissue, QRWs identify key driver genes in malignant cells that are overlooked by classical random walks. Our findings suggest that quantum random walks offer a promising alternative to classical approaches, with improved sensitivity to network structure and better performance in identifying biologically relevant features. This highlights their potential in advancing network medicine and systems biology.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 20:25:52 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:46:04 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Dubovitskii",
"Viacheslav",
""
],
[
"Bose",
"Aritra",
""
],
[
"Utro",
"Filippo",
""
],
[
"Parida",
"Laxmi",
""
]
] |
2506.06561
|
Ho Yin Ng
|
Ho Yin 'Sam' Ng, Ting-Yao Hsu, Aashish Anantha Ramakrishnan, Branislav Kveton, Nedim Lipka, Franck Dernoncourt, Dongwon Lee, Tong Yu, Sungchul Kim, Ryan A. Rossi, Ting-Hao 'Kenneth' Huang
|
LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles
|
Accepted to EMNLP 2025 Findings. The LaMP-CAP dataset is publicly available at: https://github.com/Crowd-AI-Lab/lamp-cap
| null | null | null |
cs.CL cs.AI cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Figure captions are crucial for helping readers understand and remember a figure's key message. Many models have been developed to generate these captions, helping authors compose better quality captions more easily. Yet, authors almost always need to revise generic AI-generated captions to match their writing style and the domain's style, highlighting the need for personalization. Despite language models' personalization (LaMP) advances, these technologies often focus on text-only settings and rarely address scenarios where both inputs and profiles are multimodal. This paper introduces LaMP-Cap, a dataset for personalized figure caption generation with multimodal figure profiles. For each target figure, LaMP-Cap provides not only the needed inputs, such as figure images, but also up to three other figures from the same document--each with its image, caption, and figure-mentioning paragraphs--as a profile to characterize the context. Experiments with four LLMs show that using profile information consistently helps generate captions closer to the original author-written ones. Ablation studies reveal that images in the profile are more helpful than figure-mentioning paragraphs, highlighting the advantage of using multimodal profiles over text-only ones.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 22:16:16 GMT"
},
{
"version": "v2",
"created": "Tue, 17 Jun 2025 20:40:28 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 02:00:09 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ng",
"Ho Yin 'Sam'",
""
],
[
"Hsu",
"Ting-Yao",
""
],
[
"Ramakrishnan",
"Aashish Anantha",
""
],
[
"Kveton",
"Branislav",
""
],
[
"Lipka",
"Nedim",
""
],
[
"Dernoncourt",
"Franck",
""
],
[
"Lee",
"Dongwon",
""
],
[
"Yu",
"Tong",
""
],
[
"Kim",
"Sungchul",
""
],
[
"Rossi",
"Ryan A.",
""
],
[
"Huang",
"Ting-Hao 'Kenneth'",
""
]
] |
2506.07008
|
Fatemeh Pourahmadian
|
Fatemeh Pourahmadian and Yang Xu
|
Deep regularization networks for inverse problems with noisy operators
| null | null | null | null |
math.NA cs.AI cs.NA eess.SP
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
A supervised learning approach is proposed for regularization of large inverse problems where the main operator is built from noisy data. This is germane to superresolution imaging via the sampling indicators of the inverse scattering theory. We aim to accelerate the spatiotemporal regularization process for this class of inverse problems to enable real-time imaging. In this approach, a neural operator maps each pattern on the right-hand side of the scattering equation to its affiliated regularization parameter. The network is trained in two steps which entails: (1) training on low-resolution regularization maps furnished by the Morozov discrepancy principle with nonoptimal thresholds, and (2) optimizing network predictions through minimization of the Tikhonov loss function regulated by the validation loss. Step 2 allows for tailoring of the approximate maps of Step 1 toward construction of higher quality images. This approach enables direct learning from test data and dispenses with the need for a-priori knowledge of the optimal regularization maps. The network, trained on low-resolution data, quickly generates dense regularization maps for high-resolution imaging. We highlight the importance of the training loss function on the network's generalizability. In particular, we demonstrate that networks informed by the logic of discrepancy principle lead to images of higher contrast. In this case, the training process involves many-objective optimization. We propose a new method to adaptively select the appropriate loss weights during training without requiring an additional optimization process. The proposed approach is synthetically examined for imaging damage evolution in an elastic plate. The results indicate that the discrepancy-informed regularization networks not only accelerate the imaging process, but also remarkably enhance the image quality in complex environments.
|
[
{
"version": "v1",
"created": "Sun, 08 Jun 2025 06:19:18 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 01:09:32 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Pourahmadian",
"Fatemeh",
""
],
[
"Xu",
"Yang",
""
]
] |
2506.07960
|
Ari Vesalainen
|
Ari Vesalainen, Jenna Kanerva, Aida Nitsch, Kiia Korsu, Ilari Larkiola, Laura Ruotsalainen and Filip Ginter
|
Creating a Historical Migration Dataset from Finnish Church Records, 1800-1920
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
This article presents a large-scale effort to create a structured dataset of internal migration in Finland between 1800 and 1920 using digitized church moving records. These records, maintained by Evangelical-Lutheran parishes, document the migration of individuals and families and offer a valuable source for studying historical demographic patterns. The dataset includes over six million entries extracted from approximately 200,000 images of handwritten migration records.
The data extraction process was automated using a deep learning pipeline that included layout analysis, table detection, cell classification, and handwriting recognition. The complete pipeline was applied to all images, resulting in a structured dataset suitable for research.
The dataset can be used to study internal migration, urbanization, and family migration, and the spread of disease in preindustrial Finland. A case study from the Elim\"aki parish shows how local migration histories can be reconstructed. The work demonstrates how large volumes of handwritten archival material can be transformed into structured data to support historical and demographic research.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 17:32:55 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Jul 2025 13:29:34 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 06:40:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Vesalainen",
"Ari",
""
],
[
"Kanerva",
"Jenna",
""
],
[
"Nitsch",
"Aida",
""
],
[
"Korsu",
"Kiia",
""
],
[
"Larkiola",
"Ilari",
""
],
[
"Ruotsalainen",
"Laura",
""
],
[
"Ginter",
"Filip",
""
]
] |
2506.08476
|
Jeongmin Kim
|
Hyungshick Park, Bong June Sung and Jeongmin Kim
|
Bridging Electrostatic Screening and Ion Transport in Lithium Salt-Doped Ionic Liquids
|
11 pages, 5 figures
| null | null | null |
cond-mat.soft physics.chem-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Alkali salt-doped ionic liquids are emerging as promising electrolyte systems for energy applications, owing to their excellent interfacial stability. To address their limited ionic conductivity, various strategies have been proposed, including modifying the ion solvation environment and enhancing the transport of selected ions (e.g., Li$^+$). Despite the pivotal role of electrostatic interactions in determining key physicochemical properties, their influence on ion transport in such systems has received relatively little attention. In this work, we investigate the connection between ion transport and electrostatic screening using atomistic molecular dynamics simulations of 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide ([pyr$_{14}$][TFSI]) doped with lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) at molar fractions x$_{LiTFSI}$ $\le$ 0.3. We find that the charge-charge and density-density correlation functions exhibit oscillatory exponential decay, indicating that LiTFSI doped [pyr$_{14}$][TFSI] is a charge- and mass-dense system. The electrostatic screening length decreases with increasing LiTFSI concentration, whereas the decay length of the density-density correlation functions remains nearly unchanged. Notably, we find that the x$_{LiTFSI}$-sensitive screening length serves as a central length scale for disentangling species-specific contributions of ion pairs to collective ion transport upon LiTFSI doping. This framework provides a unifying perspective on the interplay between structure and transport in ionic liquid systems.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 06:04:42 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:12:37 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Park",
"Hyungshick",
""
],
[
"Sung",
"Bong June",
""
],
[
"Kim",
"Jeongmin",
""
]
] |
2506.08768
|
Aisha Alansari
|
Ahmed Hasanaath, Aisha Alansari, Ahmed Ashraf, Chafik Salmane, Hamzah Luqman, Saad Ezzini
|
AraReasoner: Evaluating Reasoning-Based LLMs for Arabic NLP
| null | null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) have shown remarkable progress in reasoning abilities and general natural language processing (NLP) tasks, yet their performance on Arabic data, characterized by rich morphology, diverse dialects, and complex script, remains underexplored. This paper presents a comprehensive benchmarking study of multiple reasoning-focused LLMs, with a special emphasis on the newly introduced DeepSeek models, across a suite of fifteen Arabic NLP tasks. We experiment with various strategies, including zero-shot, few-shot, and fine-tuning. This allows us to systematically evaluate performance on datasets covering a range of applications to examine their capacity for linguistic reasoning under different levels of complexity. Our experiments reveal several key findings. First, carefully selecting just three in-context examples delivers an average uplift of over 13 F1 points on classification tasks-boosting sentiment analysis from 35.3% to 87.5% and paraphrase detection from 56.1% to 87.0%. Second, reasoning-focused DeepSeek architectures outperform a strong GPT o4-mini baseline by an average of 12 F1 points on complex inference tasks in the zero-shot setting. Third, LoRA-based fine-tuning yields up to an additional 8 points in F1 and BLEU compared to equivalent increases in model scale. The code is available at https://anonymous.4open.science/r/AraReasoner41299
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 13:10:31 GMT"
},
{
"version": "v2",
"created": "Wed, 11 Jun 2025 09:00:02 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 12:16:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hasanaath",
"Ahmed",
""
],
[
"Alansari",
"Aisha",
""
],
[
"Ashraf",
"Ahmed",
""
],
[
"Salmane",
"Chafik",
""
],
[
"Luqman",
"Hamzah",
""
],
[
"Ezzini",
"Saad",
""
]
] |
2506.09128
|
Peter Taylor
|
Peter L. Taylor
|
Computing Nonlinear Power Spectra Across Dynamical Dark Energy Model Space with Neural ODEs
|
Published in the Open Journal of Astrophysics. 6 pages, 4 figures
| null | null | null |
astro-ph.CO
|
http://creativecommons.org/licenses/by/4.0/
|
I show how to compute the nonlinear power spectrum across the entire $w(z)$ dynamical dark energy model space. Using synthetic $\Lambda$CDM data, I train a neural ordinary differential equation (ODE) to infer the evolution of the nonlinear matter power spectrum as a function of the background expansion and mean matter density across $\sim$$9 {\rm \ Gyr}$ of cosmic evolution. After training, the model generalises to {\it any} dynamical dark energy model parameterised by $w(z)$. With little optimisation, the neural ODE is accurate to within $4\%$ up to k = $5 \ h {\rm Mpc}^{-1}$. Unlike simulation rescaling methods, neural ODEs naturally extend to summary statistics beyond the power spectrum that are sensitive to the growth history.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 18:00:04 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 12:49:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Taylor",
"Peter L.",
""
]
] |
2506.09157
|
Mainak Mukhopadhyay
|
Mainak Mukhopadhyay, Shigeo S. Kimura
|
Electromagnetic signatures from pulsar remnants of binary neutron star mergers: prospects for unique identification using multi-wavelength signatures
|
16 pages, 10 figures; published version
|
Astrophys. J. Lett. 989 (2025) 2, L41
|
10.3847/2041-8213/adf285
| null |
astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Binary neutron star (BNS) mergers can result in the formation of long-lived magnetar remnants which can enhance neutrino and electromagnetic (EM) emissions. In this work, we study the resulting multi-wavelength EM emissions and the prospects of their detectability in the current and upcoming EM telescopes. We model the pulsar-wind neubla system where the long-lived pulsar with dipolar magnetic fields of $10^{13} - 10^{15}$ G (magnetar) spins down and is surrounded by an outward expanding nebula and kilonova ejecta. Although at early times post the merger, the EM signatures are unobservable due to heavy attenuation, they become observable on timescales of $\mathcal{O}(1 - 10)$ days after the merger. We find that the survey and follow-up observations have horizon distances $\gtrsim 1\ \rm Gpc$ for most of the wavebands and conclude that the detection prospects for such long-lived remnants in the electromagnetic channel are promising. This is of crucial importance for multi-messenger observations from BNS mergers to constrain the physical parameters of the remnants. Furthermore, we highlight how observations across the electromagnetic band can uniquely identify magnetar-powered transients resulting from BNS mergers and establish concrete associations of the detected gravitational wave signatures with such transients.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 18:11:53 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:55:04 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mukhopadhyay",
"Mainak",
""
],
[
"Kimura",
"Shigeo S.",
""
]
] |
2506.09558
|
Eleni Gkovedarou
|
Eleni Gkovedarou, Joke Daems, Luna De Bruyne
|
Gender Bias in English-to-Greek Machine Translation
|
Accepted at GITT 2025 (MT Summit)
|
https://aclanthology.org/2025.gitt-1.2/#
| null | null |
cs.CL
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
As the demand for inclusive language increases, concern has grown over the susceptibility of machine translation (MT) systems to reinforce gender stereotypes. This study investigates gender bias in two commercial MT systems, Google Translate and DeepL, focusing on the understudied English-to-Greek language pair. We address three aspects of gender bias: i) male bias, ii) occupational stereotyping, and iii) errors in anti-stereotypical translations. Additionally, we explore the potential of prompted GPT-4o as a bias mitigation tool that provides both gender-explicit and gender-neutral alternatives when necessary. To achieve this, we introduce GendEL, a manually crafted bilingual dataset of 240 gender-ambiguous and unambiguous sentences that feature stereotypical occupational nouns and adjectives. We find persistent gender bias in translations by both MT systems; while they perform well in cases where gender is explicitly defined, with DeepL outperforming both Google Translate and GPT-4o in feminine gender-unambiguous sentences, they are far from producing gender-inclusive or neutral translations when the gender is unspecified. GPT-4o shows promise, generating appropriate gendered and neutral alternatives for most ambiguous cases, though residual biases remain evident.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 09:44:12 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Gkovedarou",
"Eleni",
""
],
[
"Daems",
"Joke",
""
],
[
"De Bruyne",
"Luna",
""
]
] |
2506.09964
|
Bruno Anghinoni
|
B. Anghinoni, L. C. Malacarne, G. V. B. Lukasievicz, B. Lendl, M. L. Baesso and N. G. C. Astrath
|
Optical force density and surface displacements in transparent dielectrics due to non-ionizing sub-picosecond laser excitation
| null | null |
10.1103/9lfj-vc1n
| null |
physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
The optical force density acting in transparent dielectric media due to short laser excitation is theoretically analyzed. For typical laser pulses with picosecond duration, the momentum component of the optical force becomes of the same order of magnitude as the stress component, enhancing the overall optomechanical effects. Simulations of the optically-induced surface displacements in fused silica glass are also presented. With careful choice of realistic simulation parameters, dispersive and nonlinear effects were shown to be suppressed and displacements of 50 pm were found for single-pulse excitation with 100 fs duration, where the momentum component of the force is dominant. A possibility of measuring these displacements with piezo-electric detection is also discussed, providing a way to attain spatiotemporal characterization of the optical momentum force in continuum media. Such a description would significantly advance our current knowledge on the behavior of optical forces, potentially contributing to more versatile optomechanical applications, while also clarifying the ongoing Abraham-Minkowski fundamental problem on the momentum transferred by light.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 17:41:47 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Anghinoni",
"B.",
""
],
[
"Malacarne",
"L. C.",
""
],
[
"Lukasievicz",
"G. V. B.",
""
],
[
"Lendl",
"B.",
""
],
[
"Baesso",
"M. L.",
""
],
[
"Astrath",
"N. G. C.",
""
]
] |
2506.10024
|
Elena Sofia Ruzzetti
|
Elena Sofia Ruzzetti, Giancarlo A. Xompero, Davide Venditti, Fabio Massimo Zanzotto
|
Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models
|
To be published at ACL 2025 (Main)
| null |
10.18653/v1/2025.acl-long.810
| null |
cs.CR cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) memorize, and thus, among huge amounts of uncontrolled data, may memorize Personally Identifiable Information (PII), which should not be stored and, consequently, not leaked. In this paper, we introduce Private Memorization Editing (PME), an approach for preventing private data leakage that turns an apparent limitation, that is, the LLMs' memorization ability, into a powerful privacy defense strategy. While attacks against LLMs have been performed exploiting previous knowledge regarding their training data, our approach aims to exploit the same kind of knowledge in order to make a model more robust. We detect a memorized PII and then mitigate the memorization of PII by editing a model knowledge of its training data. We verify that our procedure does not affect the underlying language model while making it more robust against privacy Training Data Extraction attacks. We demonstrate that PME can effectively reduce the number of leaked PII in a number of configurations, in some cases even reducing the accuracy of the privacy attacks to zero.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 17:57:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ruzzetti",
"Elena Sofia",
""
],
[
"Xompero",
"Giancarlo A.",
""
],
[
"Venditti",
"Davide",
""
],
[
"Zanzotto",
"Fabio Massimo",
""
]
] |
2506.10109
|
Haohua Deng
|
Haohua Deng
|
On the generalized toroidal completion of period mappings
|
3rd Version: 30 pages. The general existence of CMCX is set back to a conjecture. New case for this conjecture is provided. The application on the weakly special subvarieties will be moved to a new paper
| null | null | null |
math.AG
|
http://creativecommons.org/licenses/by/4.0/
|
Given a period map defined over a quasi-projective variety, we construct a completion with rich geometric and Hodge-theoretic meaning. This result may be regarded as an analog of Mumford's toroidal compactification for locally symmetric Hodge varieties as well as a realizable alternative of Kato--Nakayama--Usui's construction.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 18:49:24 GMT"
},
{
"version": "v2",
"created": "Tue, 01 Jul 2025 04:54:32 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 04:53:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Deng",
"Haohua",
""
]
] |
2506.10215
|
Ole S\"onnerborn
|
Ole S\"onnerborn
|
Isoholonomic inequalities and speed limits for cyclic systems
|
17 pages, 6 figures, comments are welcome
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Quantum speed limits set fundamental lower bounds on the time required for a quantum system to evolve between states. Traditional bounds, such as those by Mandelstam-Tamm and Margolus-Levitin, rely on state distinguishability and become trivial for cyclic evolutions where the initial and final states coincide. In this work, we explore an alternative approach based on isoholonomic inequalities, which bound the length of closed state space trajectories in terms of their holonomy. Building on a gauge-theoretic framework for mixed state geometric phases, we extend the concept of isoholonomic inequalities to closed curves of isospectral and isodegenerate density operators. This allows us to derive a new quantum speed limit that remains nontrivial for cyclic evolutions. Our results reveal deep connections between the temporal behavior of cyclic quantum systems and holonomy.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 22:21:20 GMT"
},
{
"version": "v2",
"created": "Sun, 13 Jul 2025 10:48:37 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 09:27:27 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Sönnerborn",
"Ole",
""
]
] |
2506.10589
|
Nadine Ehmann
|
Nadine Ehmann, Matthias K\"ohler, Frank Allg\"ower
|
Transient performance of MPC for tracking without terminal constraints
| null |
IEEE Control Systems Letters, vol. 9, pp. 2049-2054, 2025
|
10.1109/LCSYS.2025.3585945
| null |
eess.SY cs.SY math.OC
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Model predictive control (MPC) for tracking is a recently introduced approach, which extends standard MPC formulations by incorporating an artificial reference as an additional optimization variable, in order to track external and potentially time-varying references. In this work, we analyze the performance of such an MPC for tracking scheme without a terminal cost and terminal constraints. We derive a transient performance estimate, i.e. a bound on the closed-loop performance over an arbitrary time interval, yielding insights on how to select the scheme's parameters for performance. Furthermore, we show that in the asymptotic case, where the prediction horizon and observed time interval tend to infinity, the closed-loop solution of MPC for tracking recovers the infinite horizon optimal solution.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 11:29:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ehmann",
"Nadine",
""
],
[
"Köhler",
"Matthias",
""
],
[
"Allgöwer",
"Frank",
""
]
] |
2506.10648
|
Weiyu Shen
|
Weiyu Shen, Rodolfo Ostilla-M\'onico and Xiaojue Zhu
|
Vortex-magnetic competition and regime transitions in antiparallel flux tubes
| null | null | null | null |
physics.flu-dyn astro-ph.SR physics.plasm-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Vortex-magnetic interactions shape magnetohydrodynamic (MHD) turbulence, influencing energy transfer in astrophysical, geophysical, and industrial systems. On the Sun, granular-scale vortex flows couple strongly with magnetic fields, channeling energy into the corona. At high Reynolds numbers, vorticity and magnetic fields are nearly frozen into the charged fluid, and MHD flows emerge from the Lorentz force mediated interactions between coherent vortex structures in matter and the field. To probe this competition in a controlled setting, we revisit the canonical problem of two antiparallel flux tubes. By varying the magnetic flux threading each tube--and thus sweeping the interaction parameter $N_i$, which gauges Lorentz-to-inertial force balance--we uncover three distinct regimes: vortex-dominated joint reconnection, instability-triggered cascade, and Lorentz-induced vortex disruption. At low $N_i$, classical vortex dynamics dominate, driving joint vortex-magnetic reconnection and amplifying magnetic energy via a dynamo effect. At moderate $N_i$, the system oscillates between vorticity-driven attraction and magnetic damping, triggering instabilities and nonlinear interactions that spawn secondary filaments and drive an energy cascade. At high $N_i$, Lorentz forces suppress vortex interactions, aligning the tubes axially while disrupting vortex cores and rapidly converting magnetic to kinetic energy. These findings reveal how the inertial-Lorentz balance governs energy transfer and coherent structure formation in MHD turbulence, offering insight into vortex-magnetic coevolution in astrophysical plasmas.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 12:38:31 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:01:57 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Shen",
"Weiyu",
""
],
[
"Ostilla-Mónico",
"Rodolfo",
""
],
[
"Zhu",
"Xiaojue",
""
]
] |
2506.10701
|
Simon Elias Schrader
|
Simon Elias Schrader, H{\aa}kon Emil Kristiansen, Thomas Bondo Pedersen, Simen Kvaal
|
Time-dependent Gaussian basis sets for many-body systems using Rothe's method: A mean-field study
|
22 pages, 10 figures
| null |
10.1021/acs.jctc.5c00970
| null |
physics.chem-ph physics.comp-ph
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
A challenge in modeling time-dependent strong-field processes such as high-harmonic generation for many-body systems, is how to effectively represent the electronic continuum. We apply Rothe's method to the time-dependent Hartree-Fock (TDHF) and density functional theory (TDDFT) equations of motion for the orbitals, which reformulates them as an optimization problem. We show that thawed, complex-valued Gaussian basis sets can be propagated efficiently for these orbital-based approaches, removing the need for grids. In particular, we illustrate that qualitatively correct results can often be obtained by using just a few fully flexible Gaussians that describe the unbound dynamics for both TDHF and TDDFT. Grid calculations can be reproduced quantitatively using $30$--$100$ Gaussians for intensities up to $4\times10^{14}$ W/cm$^2$ for the one-dimensional molecular systems considered in this work.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 13:51:29 GMT"
},
{
"version": "v2",
"created": "Mon, 11 Aug 2025 06:49:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Schrader",
"Simon Elias",
""
],
[
"Kristiansen",
"Håkon Emil",
""
],
[
"Pedersen",
"Thomas Bondo",
""
],
[
"Kvaal",
"Simen",
""
]
] |
2506.10913
|
Ashley Samuelson
|
Ashley Samuelson, Andrew K. Hirsch, Ethan Cecchetti
|
Choreographic Quick Changes: First-Class Location (Set) Polymorphism
|
36 pages, 11 figures
| null | null | null |
cs.PL
|
http://creativecommons.org/licenses/by/4.0/
|
Choreographic programming is a promising new paradigm for programming concurrent systems where a developer writes a single centralized program that compiles to individual programs for each node. Existing choreographic languages, however, lack critical features integral to modern systems, like the ability of one node to dynamically compute who should perform a computation and send that decision to others. This work addresses this gap with $\lambda_{QC}$, the first typed choreographic language with \emph{first class process names} and polymorphism over both types and (sets of) locations. $\lambda_{QC}$ also improves expressive power over previous work by supporting algebraic and recursive data types as well as multiply-located values. We formalize and mechanically verify our results in Rocq, including the standard choreographic guarantee of deadlock freedom.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 17:27:48 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 21:50:16 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Samuelson",
"Ashley",
""
],
[
"Hirsch",
"Andrew K.",
""
],
[
"Cecchetti",
"Ethan",
""
]
] |
2506.12263
|
Hui Wei
|
Hui Wei, Dong Yoon Lee, Shubham Rohal, Zhizhang Hu, Ryan Rossi, Shiwei Fang, Shijia Pan
|
A Survey of Foundation Models for IoT: Taxonomy and Criteria-Based Analysis
|
Accepted by CCF Transactions on Pervasive Computing and Interaction
| null | null | null |
cs.LG cs.AI cs.SY eess.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Foundation models have gained growing interest in the IoT domain due to their reduced reliance on labeled data and strong generalizability across tasks, which address key limitations of traditional machine learning approaches. However, most existing foundation model based methods are developed for specific IoT tasks, making it difficult to compare approaches across IoT domains and limiting guidance for applying them to new tasks. This survey aims to bridge this gap by providing a comprehensive overview of current methodologies and organizing them around four shared performance objectives by different domains: efficiency, context-awareness, safety, and security & privacy. For each objective, we review representative works, summarize commonly-used techniques and evaluation metrics. This objective-centric organization enables meaningful cross-domain comparisons and offers practical insights for selecting and designing foundation model based solutions for new IoT tasks. We conclude with key directions for future research to guide both practitioners and researchers in advancing the use of foundation models in IoT applications.
|
[
{
"version": "v1",
"created": "Fri, 13 Jun 2025 22:28:55 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 03:53:23 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wei",
"Hui",
""
],
[
"Lee",
"Dong Yoon",
""
],
[
"Rohal",
"Shubham",
""
],
[
"Hu",
"Zhizhang",
""
],
[
"Rossi",
"Ryan",
""
],
[
"Fang",
"Shiwei",
""
],
[
"Pan",
"Shijia",
""
]
] |
2506.13158
|
Rickmoy Samanta
|
Aswathy K R, Udaya Maurya, Surya Teja Gavva and Rickmoy Samanta
|
Dynamics of Vortex Clusters on a Torus
| null | null | null | null |
physics.flu-dyn cond-mat.quant-gas cond-mat.soft hep-th math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the collective dynamics of multivortex assemblies in a two dimensional (2D) toroidal fluid film of distinct curvature and topology. The incompressible and inviscid nature of the fluid allows a Hamiltonian description of the vortices, along with a self-force of geometric origin, arising from the standard Kirchhoff-Routh regularization procedure. The Hamiltonian dynamics is constructed in terms of $q$-digamma functions $\Psi_q(z)$, closely related to the Schottky-Klein prime function known to arise in multiply connected domains. We show the fundamental motion of the two-vortex system and identify five classes of geodesics on the torus for the special case of a vortex dipole, along with subtle distinctions from vortices in quantum superfluids. In multivortex assemblies, we observe that a randomly initialized cluster of vortices of the same sign and strength (chiral cluster) remains geometrically confined on the torus, while undergoing an overall drift along the toroidal direction, exhibiting collective dynamics. A cluster of fast and slow vortices also show the collective toroidal drift, with the fast ones predominantly occupying the core region and the slow ones expelled to the periphery of the revolving cluster. Vortex clusters of mixed sign but zero net circulation (achiral cluster) show unconfined dynamics and scatter all over the surface of the torus. A chiral cluster with an impurity in the form of a single vortex of opposite sign also show similar behavior as a pure chiral cluster, with occasional ``jets" of dipoles leaving and re-entering the revolving cluster. The work serves as a step towards analysis of vortex clusters in models that incorporate harmonic velocities in the Hodge decomposition.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 07:10:59 GMT"
},
{
"version": "v2",
"created": "Sat, 26 Jul 2025 07:16:07 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 17:25:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"R",
"Aswathy K",
""
],
[
"Maurya",
"Udaya",
""
],
[
"Gavva",
"Surya Teja",
""
],
[
"Samanta",
"Rickmoy",
""
]
] |
2506.14872
|
Hideki Maeda
|
Hideki Maeda, Cristian Martinez
|
Exact plane symmetric black bounce with a perfect-fluid exterior obeying a linear equation of state
|
14 pages, 2 fiures, 4 tables; v2, final version published in Physical Review D
|
Phys.Rev.D 112 (2025) 4, 044021
|
10.1103/lt5f-gqmw
| null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate an exact two-parameter family of plane symmetric solutions admitting a hypersurface-orthogonal Killing vector in general relativity with a perfect fluid obeying a linear equation of state $p=\chi\rho$ in $n(\ge 4)$ dimensions, obtained by Gamboa in 2012. The Gamboa solution is identical to the topological Schwarzschild-Tangherlini-(anti-)de~Sitter $\Lambda$-vacuum solution for $\chi=-1$ and admits a nondegenerate Killing horizon only for $\chi=-1$ and $\chi\in[-1/3,0)$. We identify all possible regular attachments of two Gamboa solutions for $\chi\in[-1/3,0)$ at the Killing horizon without a lightlike thin shell, where $\chi$ may have different values on each side of the horizon. We also present the maximal extension of the static and asymptotically topological Schwarzschild-Tangherlini Gamboa solution, realized only for $\chi\in(-(n-3)/(3n-5),0)$, under the assumption that the value of $\chi$ is unchanged in the extended dynamical region beyond the horizon. The maximally extended spacetime describes either (i) a globally regular black bounce whose Killing horizon coincides with a bounce null hypersurface or (ii) a black hole with a spacelike curvature singularity inside the horizon. The matter field inside the horizon is not a perfect fluid but rather an anisotropic fluid that can be interpreted as a spacelike (tachyonic) perfect fluid. A fine-tuning of the parameters is unnecessary for the black bounce, but the null energy condition is violated everywhere except on the horizon. In the black-bounce (black-hole) case, the metric in the regular coordinate system is $C^\infty$ only for $\chi=-1/(1+2N)$ with odd (even) $N$ satisfying $N>(n-1)/(n-3)$, and if one of the parameters in the extended region is fine-tuned.
|
[
{
"version": "v1",
"created": "Tue, 17 Jun 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 22:38:31 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Maeda",
"Hideki",
""
],
[
"Martinez",
"Cristian",
""
]
] |
2506.15784
|
Pablo Antonio Moreno Casares
|
Stepan Fomichev and Pablo A. M. Casares and Jay Soni and Utkarsh Azad and Alexander Kunitsa and Arne-Christian Voigt and Jonathan E. Mueller and Juan Miguel Arrazola
|
Fast simulations of X-ray absorption spectroscopy for battery materials on a quantum computer
|
29 pages. The changes reflect a more accurate treatment of the impact of qubit rotation inaccuracies, and a new appendix has been added to analyze this choice
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
X-ray absorption spectroscopy (XAS) is a leading technique for understanding structural changes in advanced battery materials such as lithium-excess cathodes. However, extracting critical information like oxidation states from the experimental spectra requires expensive and time-consuming simulations. Building upon a recent proposal to simulate XAS using quantum computers, this work proposes a highly-optimized implementation of the time-domain algorithm for X-ray absorption. Among a host of improvements to Hamiltonian representation, circuit implementation, and measurement strategies, three optimizations are key to the efficiency of the algorithm. The first is the use of product formulas with the compressed double factorized form of the Hamiltonian. The second is recognizing that for spectroscopy applications, it is sufficient to control the error in the eigenvalues of the (approximate) Hamiltonian being implemented by the product formula, rather than the generic error on the full time evolution operator. Using perturbation theory to estimate this eigenvalue error, we find that significantly fewer Trotter steps are needed than expected from the time evolution error bound. The third is the choice of an optimized distribution of samples that takes advantage of the exponentially decaying Lorentzian kernel. Through constant factor resource estimates, we show that a challenging model Li$_4$Mn$_2$O cluster system with 18 spatial orbitals and 22 electrons in the active space can be simulated with 100 logical qubits and less than $4 \times 10^8$ T gates per circuit. Finally, the algorithm is implemented on a simulator, and the reconstructed spectrum is verified against a classical computational reference. The low cost of our algorithm makes it attractive to use on fault-tolerant quantum devices to accelerate the development and commercialization of high-capacity battery cathodes.
|
[
{
"version": "v1",
"created": "Wed, 18 Jun 2025 18:00:29 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 18:56:21 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fomichev",
"Stepan",
""
],
[
"Casares",
"Pablo A. M.",
""
],
[
"Soni",
"Jay",
""
],
[
"Azad",
"Utkarsh",
""
],
[
"Kunitsa",
"Alexander",
""
],
[
"Voigt",
"Arne-Christian",
""
],
[
"Mueller",
"Jonathan E.",
""
],
[
"Arrazola",
"Juan Miguel",
""
]
] |
2506.15928
|
Myke C. Cohen
|
Myke C. Cohen, Zhe Su, Hsien-Te Kao, Daniel Nguyen, Spencer Lynch, Maarten Sap, Svitlana Volkova
|
Exploring Big Five Personality and AI Capability Effects in LLM-Simulated Negotiation Dialogues
|
Presented at the KDD 2025 Workshop on Evaluation and Trustworthiness of Agentic and Generative AI Models under the title "Evaluating the LLM-simulated Impacts of Big Five Personality Traits and AI Capabilities on Social Negotiations" (https://kdd-eval-workshop.github.io/genai-evaluation-kdd2025/assets/papers/Submission%2036.pdf)
| null | null | null |
cs.AI cs.CL cs.HC
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This paper presents an evaluation framework for agentic AI systems in mission-critical negotiation contexts, addressing the need for AI agents that can adapt to diverse human operators and stakeholders. Using Sotopia as a simulation testbed, we present two experiments that systematically evaluated how personality traits and AI agent characteristics influence LLM-simulated social negotiation outcomes--a capability essential for a variety of applications involving cross-team coordination and civil-military interactions. Experiment 1 employs causal discovery methods to measure how personality traits impact price bargaining negotiations, through which we found that Agreeableness and Extraversion significantly affect believability, goal achievement, and knowledge acquisition outcomes. Sociocognitive lexical measures extracted from team communications detected fine-grained differences in agents' empathic communication, moral foundations, and opinion patterns, providing actionable insights for agentic AI systems that must operate reliably in high-stakes operational scenarios. Experiment 2 evaluates human-AI job negotiations by manipulating both simulated human personality and AI system characteristics, specifically transparency, competence, adaptability, demonstrating how AI agent trustworthiness impact mission effectiveness. These findings establish a repeatable evaluation methodology for experimenting with AI agent reliability across diverse operator personalities and human-agent team dynamics, directly supporting operational requirements for reliable AI systems. Our work advances the evaluation of agentic AI workflows by moving beyond standard performance metrics to incorporate social dynamics essential for mission success in complex operations.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 00:14:56 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Jun 2025 23:42:18 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 19:36:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Cohen",
"Myke C.",
""
],
[
"Su",
"Zhe",
""
],
[
"Kao",
"Hsien-Te",
""
],
[
"Nguyen",
"Daniel",
""
],
[
"Lynch",
"Spencer",
""
],
[
"Sap",
"Maarten",
""
],
[
"Volkova",
"Svitlana",
""
]
] |
2506.16002
|
Kazuma Yokota
|
Kazuma Yokota, Masato Itami, Shin-ichi Sasa
|
Unifying renormalized and bare viscosity in two-dimensional molecular dynamics simulations
|
8+3 pages, 7 figures
| null | null | null |
cond-mat.stat-mech
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Fluctuating hydrodynamics provides a framework connecting mesoscopic fluctuations with macroscopic transport behavior. To bridge mesoscopic and macroscopic transport from microscopic dynamics, we introduce a wavenumber-dependent viscosity, defined via the equilibrium correlation of time-averaged Fourier components of the fine-grained shear stress field, as a finite-wavevector extension of the Green--Kubo formula. Two-dimensional molecular dynamics simulations reveal its small-wavenumber divergence characteristic of the renormalized viscosity, while its large-wavenumber behavior determines the bare viscosity, thereby establishing a link between mesoscopic and macroscopic transport based on microscopic dynamics.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 03:47:25 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:00:53 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yokota",
"Kazuma",
""
],
[
"Itami",
"Masato",
""
],
[
"Sasa",
"Shin-ichi",
""
]
] |
2506.16244
|
Alejandro D\'iaz-Caro
|
Alejandro D\'iaz-Caro and Nicolas A. Monzon
|
A Quantum-Control Lambda-Calculus with Multiple Measurement Bases
|
This includes the appendix that was omitted from the APLAS 2025 publication
| null | null | null |
cs.LO quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce Lambda-SX, a typed quantum lambda-calculus that supports multiple measurement bases. By tracking duplicability relative to arbitrary bases within the type system, Lambda-SX enables more flexible control and compositional reasoning about measurements. We formalise its syntax, typing rules, subtyping, and operational semantics, and establish its key meta-theoretical properties. This proof-of-concept shows that support for multiple bases can be coherently integrated into the type discipline of quantum programming languages.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 11:59:39 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 10:04:22 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Díaz-Caro",
"Alejandro",
""
],
[
"Monzon",
"Nicolas A.",
""
]
] |
2506.17383
|
Marcel Bernet
|
Marcel Bernet, Pau Ramos, Teresa Antoja, Adrian Price-Whelan, Steven L. Brunton, Tetsuro Asano, Alexandra Gir\'on-Soto
|
Dynamics of tidal spiral arms: Machine learning-assisted identification of equations and application to the Milky Way
|
13 pages, 9 figures. Accepted in Astronomy & Astrophysics
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Understanding the spiral arms of the Milky Way (MW) remains a key open question in galactic dynamics. Tidal perturbations, such as the recent passage of the Sagittarius dwarf galaxy (Sgr), could play a significant role in exciting them. We aim to analytically characterize the dynamics of tidally induced spiral arms, including their phase-space signatures. We ran idealized test-particle simulations resembling impulsive satellite impacts, and used the Sparse Identification of Non-linear Dynamics (SINDy) method to infer their governing Partial Differential Equations (PDEs). We validated the method with analytical derivations and a realistic $N$-body simulation of a MW-Sgr encounter analogue. For small perturbations, a linear system of equations was recovered with SINDy, consistent with predictions from linearised collisionless dynamics. In this case, two distinct waves wrapping at pattern speeds $\Omega \pm \kappa/m$ emerge. For large impacts, we empirically discovered a non-linear system of equations, representing a novel formulation for the dynamics of tidally induced spiral arms. For both cases, these equations describe wave properties like amplitude and pattern speed, and their shape and temporal evolution in different phase-space projections. We fit the Gaia $L_Z-V_R$ waves with the linear model, providing a reasonable fit and plausible parameters for the Sgr passage. However, the predicted amplitude ratio of the two waves is inconsistent with observations, supporting a more complex origin for this feature (e.g. multiple passages, bar, spiral arms). We merge data-driven discovery with theory to create simple, accurate models of tidal spiral arms that match simulations and provide a simple tool to fit Gaia and external galaxy data. This methodology could be extended to model complex phenomena like self-gravity and dynamical friction. (ABR)
|
[
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:35:33 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Bernet",
"Marcel",
""
],
[
"Ramos",
"Pau",
""
],
[
"Antoja",
"Teresa",
""
],
[
"Price-Whelan",
"Adrian",
""
],
[
"Brunton",
"Steven L.",
""
],
[
"Asano",
"Tetsuro",
""
],
[
"Girón-Soto",
"Alexandra",
""
]
] |
2506.17605
|
Ben Savoie
|
Ben Savoie
|
Infinitely many elliptic curves over $\mathbb{Q}(i)$ with rank 2 and $j$-invariant 1728
|
8 pages. Introduced the notion of "elliptic curves genuinely defined over Q(i)" to reframe the main result. Removed secondary results (Theorems 1.3, 1.4 from v1) to shorten the paper. Streamlined several arguments, particularly in Section 2, and corrected minor typos
| null | null | null |
math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove that there exist infinitely many elliptic curves over $\mathbb{Q}(i)$ with $j$-invariant $1728$ and rank exactly $2$ which are not obtained by base change from $\mathbb{Q}$. The rank of each such curve is determined via 2-isogeny descent, and the existence of infinitely many such curves follows from Tao's constellation theorem for Gaussian primes.
|
[
{
"version": "v1",
"created": "Sat, 21 Jun 2025 05:56:57 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:54:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Savoie",
"Ben",
""
]
] |
2506.18146
|
Yuan Li
|
Yuan Li, Gwenael Giacinti, Siming Liu
|
Discovery of diffuse $\gamma$-ray emission in the vicinity of G213.0-0.6: Supernova remnant versus massive star-forming region
|
9 pages, 5 figures. Accepted for publication in ApJ. DOI: 10.3847/1538-4357/adf209
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
We report the discovery of high-energy $\gamma$-ray emission in the vicinity of G213.0-0.6, which is debated as a supernova remnant (SNR) or an ionized hydrogen (H$_{\rm{II}}$) region. Using 16-yr Pass 8 data from Fermi Large Area Telescope (Fermi-LAT), we found three extended sources with different photon spectra in this region, which will label as SrcA, SrcB and SrcC. Among them, the $\gamma$-ray source SrcA with a log-parabola spectrum is spatially coincident with a star-forming region and several OB stars. The power-law spectra source SrcB is spatially coincident with a SNR radio shell. SrcC with a harder power-law photon spectrum is located outside of the radio shell structure. All of them are spatially coincident with a dense molecular cloud (MC) in the velocity range of 35 - 54 km s$^{-1}$. In this scenario, SrcB can be interpreted as the GeV counterpart of the SNR, and its $\gamma$-ray emission originates from the shock-cloud interaction. SrcA and SrcC originate from the escaped CRs illuminating nearby MC. For SrcA, another possibility is that the $\gamma$-ray emission originates from a young stellar cluster (YSC) associated with a star-forming region (SFR), however, the supporting evidence remains insufficient to draw a definitive conclusion.
|
[
{
"version": "v1",
"created": "Sun, 22 Jun 2025 19:27:36 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:01:15 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Li",
"Yuan",
""
],
[
"Giacinti",
"Gwenael",
""
],
[
"Liu",
"Siming",
""
]
] |
2506.18246
|
Xiangzhao Hao
|
Xiangzhao Hao, Kuan Zhu, Hongyu Guo, Haiyun Guo, Ning Jiang, Quan Lu, Ming Tang, Jinqiao Wang
|
Referring Expression Instance Retrieval and A Strong End-to-End Baseline
|
ACMMM2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Using natural language to query visual information is a fundamental need in real-world applications. Text-Image Retrieval (TIR) retrieves a target image from a gallery based on an image-level description, while Referring Expression Comprehension (REC) localizes a target object within a given image using an instance-level description. However, real-world applications often present more complex demands. Users typically query an instance-level description across a large gallery and expect to receive both relevant image and the corresponding instance location. In such scenarios, TIR struggles with fine-grained descriptions and object-level localization, while REC is limited in its ability to efficiently search large galleries and lacks an effective ranking mechanism. In this paper, we introduce a new task called \textbf{Referring Expression Instance Retrieval (REIR)}, which supports both instance-level retrieval and localization based on fine-grained referring expressions. First, we propose a large-scale benchmark for REIR, named REIRCOCO, constructed by prompting advanced vision-language models to generate high-quality referring expressions for instances in the MSCOCO and RefCOCO datasets. Second, we present a baseline method, Contrastive Language-Instance Alignment with Relation Experts (CLARE), which employs a dual-stream architecture to address REIR in an end-to-end manner. Given a referring expression, the textual branch encodes it into a query embedding. The visual branch detects candidate objects and extracts their instance-level visual features. The most similar candidate to the query is selected for bounding box prediction. CLARE is first trained on object detection and REC datasets to establish initial grounding capabilities, then optimized via Contrastive Language-Instance Alignment (CLIA) for improved retrieval across images. We will release our code and benchmark publicly.
|
[
{
"version": "v1",
"created": "Mon, 23 Jun 2025 02:28:44 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 03:38:39 GMT"
},
{
"version": "v3",
"created": "Thu, 26 Jun 2025 11:31:17 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 09:45:48 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hao",
"Xiangzhao",
""
],
[
"Zhu",
"Kuan",
""
],
[
"Guo",
"Hongyu",
""
],
[
"Guo",
"Haiyun",
""
],
[
"Jiang",
"Ning",
""
],
[
"Lu",
"Quan",
""
],
[
"Tang",
"Ming",
""
],
[
"Wang",
"Jinqiao",
""
]
] |
2506.19073
|
Francielle Vargas
|
Jackson Trager and Francielle Vargas and Diego Alves and Matteo Guida and Mikel K. Ngueajio and Flor Plaza-del-Arco and Yalda Daryanai and Farzan Karimi-Malekabadi and Ameeta Agrawal
|
MFTCXplain: A Multilingual Benchmark Dataset for Evaluating the Moral Reasoning of LLMs through Hate Speech Multi-hop Explanations
|
Findings of the Association for Computational Linguistics: EMNLP 2025; *These authors contributed equally
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Ensuring the moral reasoning capabilities of Large Language Models (LLMs) is a growing concern as these systems are used in socially sensitive tasks. Nevertheless, current evaluation benchmarks present two major shortcomings: a lack of annotations that justify moral classifications, which limits transparency and interpretability; and a predominant focus on English, which constrains the assessment of moral reasoning across diverse cultural settings. In this paper, we introduce MFTCXplain, a multilingual benchmark dataset for evaluating the moral reasoning of LLMs via hate speech multi-hop explanation using Moral Foundation Theory (MFT). The dataset comprises 3,000 tweets across Portuguese, Italian, Persian, and English, annotated with binary hate speech labels, moral categories, and text span-level rationales. Empirical results highlight a misalignment between LLM outputs and human annotations in moral reasoning tasks. While LLMs perform well in hate speech detection (F1 up to 0.836), their ability to predict moral sentiments is notably weak (F1 < 0.35). Furthermore, rationale alignment remains limited mainly in underrepresented languages. These findings show the limited capacity of current LLMs to internalize and reflect human moral reasoning.
|
[
{
"version": "v1",
"created": "Mon, 23 Jun 2025 19:44:21 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Jul 2025 15:54:38 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 23:30:40 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Trager",
"Jackson",
""
],
[
"Vargas",
"Francielle",
""
],
[
"Alves",
"Diego",
""
],
[
"Guida",
"Matteo",
""
],
[
"Ngueajio",
"Mikel K.",
""
],
[
"Plaza-del-Arco",
"Flor",
""
],
[
"Daryanai",
"Yalda",
""
],
[
"Karimi-Malekabadi",
"Farzan",
""
],
[
"Agrawal",
"Ameeta",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.