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
|
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2404.09740
|
Yutaka Hirano
|
Yutaka Hirano, Tomohiro Itogawa, Keisuke Fujii
|
Leveraging Zero-Level Distillation to Generate High-Fidelity Magic States
|
11 pages, 19 figures. This work has been submitted to the IEEE for possible publication
| null |
10.1109/QCE60285.2024.00104
| null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Magic state distillation plays an important role in universal fault-tolerant quantum computing, and its overhead is one of the major obstacles to realizing fault-tolerant quantum computers. Hence, many studies have been conducted to reduce this overhead. Among these, Litinski has provided a concrete assessment of resource-efficient distillation protocol implementations on the rotated surface code. On the other hand, recently, Itogawa et al. have proposed zero-level distillation, a distillation protocol offering very small spatial and temporal overhead to generate relatively low-fidelity magic states. While zero-level distillation offers preferable spatial and temporal overhead, it cannot directly generate high-fidelity magic states since it only reduces the logical error rate of the magic state quadratically. In this study, we evaluate the spatial and temporal overhead of two-level distillation implementations generating relatively high-fidelity magic states, including ones incorporating zero-level distillation. To this end, we introduce (0+1)-level distillation, a two-level distillation protocol which combines zero-level distillation and the 15-to-1 distillation protocol. We refine the second-level 15-to-1 implementation in it to capitalize on the small footprint of zero-level distillation. Under conditions of a physical error probability of $p_{\mathrm{phys}} = 10^{-4}$ ($10^{-3}$) and targeting an error rate for the magic state within $[5 \times 10^{-17}, 10^{-11}]$ ($[5 \times 10^{-11}, 10^{-8}]$), (0+1)-level distillation reduces the spatiotemporal overhead by more than 63% (61%) compared to the (15-to-1)$\times$(15-to-1) protocol and more than 43% (44%) compared to the (15-to-1)$\times$(20-to-4) protocol, offering a substantial efficiency gain over the traditional protocols.
|
[
{
"version": "v1",
"created": "Mon, 15 Apr 2024 12:42:29 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hirano",
"Yutaka",
""
],
[
"Itogawa",
"Tomohiro",
""
],
[
"Fujii",
"Keisuke",
""
]
] |
2404.10052
|
Patrick Christopher Fragile
|
P. Chris Fragile and Matthew Liska
|
Tilted Accretion Disks
|
29 pages, 15 figures, chapter for book on GRMHD Simulations
| null |
10.1007/978-981-97-8522-3_11
| null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this chapter, we review some of the interesting consequences that tilt between the spin axis of the black hole and angular momentum axis of the accretion disk can have on the dynamics, thermodynamics, and observational appearance of accreting systems, from precessing coronae and jets to standing nozzle shocks and quasi-periodic oscillations. We begin the chapter by examining some of the reasons tilted disks are interesting as well as present arguments for how ubiquitous they may be. We then review the existing simulation results in the literature, broadly dividing them into sections on thick disks, thin disks, and magnetically arrested disks (MADs). We finish by highlighting some of the phenomenology that is unique to tilted disk simulations and discuss how this may apply to observations.
|
[
{
"version": "v1",
"created": "Mon, 15 Apr 2024 18:00:22 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fragile",
"P. Chris",
""
],
[
"Liska",
"Matthew",
""
]
] |
2404.11916
|
Nakyeong Yang
|
Nakyeong Yang, Jiwon Moon, Junseok Kim, Yunah Jang, Kyomin Jung
|
Unplug and Play Language Models: Decomposing Experts in Language Models at Inference Time
|
accepted to CIKM 2025
| null |
10.1145/3746252.3761011
| null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Enabled by large-scale text corpora with huge parameters, pre-trained language models operate as multi-task experts using a single model architecture. However, recent studies have revealed that certain neurons play disproportionately important roles in solving specific tasks, suggesting that task-relevant substructures can be isolated and selectively activated for each task. Therefore, we introduce Decomposition of Experts (DoE), a novel framework that dynamically identifies and activates task-specific experts within a language model to reduce inference cost without sacrificing accuracy. We first define a task expert as a set of parameters that significantly influence the performance of a specific task and propose a four-step unplug-and-play process: (1) receiving a user request, (2) identifying the corresponding task expert, (3) performing inference using the expert-localized model, and (4) restoring the original model and waiting for the next task. Using attribution methods and prompt tuning, DoE isolates task-relevant neurons, minimizing computational overhead while maintaining task performance. We assume a setting where a language model receives user requests from five widely used natural language understanding benchmarks, processing one task at a time. In this setup, we demonstrate that DoE achieves up to a x1.73 inference speed-up with a 65% pruning rate, without compromising accuracy. Comparisons with various task expert localization methods reveal that DoE effectively identifies task experts, while ablation studies validate the importance of its components. Additionally, we analyze the effects of batch size, token count, and layer types on inference speed-up, providing practical insights for adopting DoE. The proposed framework is both practical and scalable, applicable to any transformer-based architecture, offering a robust solution for efficient task-specific inference.
|
[
{
"version": "v1",
"created": "Thu, 18 Apr 2024 05:43:50 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Oct 2024 09:01:27 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 14:04:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yang",
"Nakyeong",
""
],
[
"Moon",
"Jiwon",
""
],
[
"Kim",
"Junseok",
""
],
[
"Jang",
"Yunah",
""
],
[
"Jung",
"Kyomin",
""
]
] |
2404.12590
|
A. Feder Cooper
|
A. Feder Cooper and James Grimmelmann
|
The Files are in the Computer: Copyright, Memorization, and Generative AI
|
Forthcoming, Chicago-Kent Law Review
|
Chicago-Kent Law Review Vol. 100, 2025
| null | null |
cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
The New York Times's copyright lawsuit against OpenAI and Microsoft alleges OpenAI's GPT models have "memorized" NYT articles. Other lawsuits make similar claims. But parties, courts, and scholars disagree on what memorization is, whether it is taking place, and what its copyright implications are. These debates are clouded by ambiguities over the nature of "memorization." We attempt to bring clarity to the conversation. We draw on the technical literature to provide a firm foundation for legal discussions, providing a precise definition of memorization: a model has "memorized" a piece of training data when (1) it is possible to reconstruct from the model (2) a near-exact copy of (3) a substantial portion of (4) that piece of training data. We distinguish memorization from "extraction" (user intentionally causes a model to generate a near-exact copy), from "regurgitation" (model generates a near-exact copy, regardless of user intentions), and from "reconstruction" (the near-exact copy can be obtained from the model by any means). Several consequences follow. (1) Not all learning is memorization. (2) Memorization occurs when a model is trained; regurgitation is a symptom not its cause. (3) A model that has memorized training data is a "copy" of that training data in the sense used by copyright. (4) A model is not like a VCR or other general-purpose copying technology; it is better at generating some types of outputs (possibly regurgitated ones) than others. (5) Memorization is not a phenomenon caused by "adversarial" users bent on extraction; it is latent in the model itself. (6) The amount of training data that a model memorizes is a consequence of choices made in training. (7) Whether or not a model that has memorized actually regurgitates depends on overall system design. In a very real sense, memorized training data is in the model--to quote Zoolander, the files are in the computer.
|
[
{
"version": "v1",
"created": "Fri, 19 Apr 2024 02:37:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Jul 2024 13:58:13 GMT"
},
{
"version": "v3",
"created": "Thu, 18 Jul 2024 22:41:58 GMT"
},
{
"version": "v4",
"created": "Mon, 11 Nov 2024 19:59:04 GMT"
},
{
"version": "v5",
"created": "Mon, 10 Mar 2025 17:10:16 GMT"
},
{
"version": "v6",
"created": "Mon, 24 Mar 2025 16:58:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Cooper",
"A. Feder",
""
],
[
"Grimmelmann",
"James",
""
]
] |
2404.13261
|
Xiao-Chuan Xu
|
Xiao-Chuan Xu, Chuan-Fu Yang and Natalia Pavlovna Bondarenko
|
On the local solvability and stability of the partial inverse problems for the non-self-adjoint Sturm-Liouville operators with a discontinuity
|
26 pages
|
J. Math. Phys. 66, 081504 (2025)
|
10.1063/5.0233344
| null |
math.SP math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this work, we study the inverse spectral problems for the Sturm-Liouville operators on [0,1] with complex coefficients and a discontinuity at $x=a\in(0,1)$. Assume that the potential on (a,1) and some parameters in the discontinuity and boundary conditions are given. We recover the potential on (0,a) and the other parameters from the eigenvalues. This is the so-called partial inverse problem. The local solvability and stability of the partial inverse problems are obtained for $a\in(0,1)$, in which the error caused by the given partial potential is considered. As a by-product, we also obtain two new uniqueness theorems for the partial inverse problem.
|
[
{
"version": "v1",
"created": "Sat, 20 Apr 2024 04:11:18 GMT"
},
{
"version": "v2",
"created": "Fri, 12 Jul 2024 08:40:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Xu",
"Xiao-Chuan",
""
],
[
"Yang",
"Chuan-Fu",
""
],
[
"Bondarenko",
"Natalia Pavlovna",
""
]
] |
2404.14204
|
Guanqiao Qu
|
Guanqiao Qu, Zheng Lin, Qian Chen, Jian Li, Fangming Liu, Xianhao Chen, Kaibin Huang
|
TrimCaching: Parameter-sharing Edge Caching for AI Model Downloading
|
18 pages, 13 figures. Part of this work has been accepted by ICDCS 2024
| null | null | null |
cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm of edge model caching. In this paper, we develop a novel model placement framework, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To tackle this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with a $\left(1-\epsilon\right)/2$-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models.
|
[
{
"version": "v1",
"created": "Mon, 22 Apr 2024 14:13:36 GMT"
},
{
"version": "v2",
"created": "Sun, 12 May 2024 12:50:42 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 11:43:48 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Qu",
"Guanqiao",
""
],
[
"Lin",
"Zheng",
""
],
[
"Chen",
"Qian",
""
],
[
"Li",
"Jian",
""
],
[
"Liu",
"Fangming",
""
],
[
"Chen",
"Xianhao",
""
],
[
"Huang",
"Kaibin",
""
]
] |
2404.14280
|
Fadi Khatib
|
Fadi Khatib, Yoni Kasten, Dror Moran, Meirav Galun and Ronen Basri
|
RESfM: Robust Deep Equivariant Structure from Motion
|
Accepted to ICLR 2025. Project page: https://robust-equivariant-sfm.github.io/
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Multiview Structure from Motion is a fundamental and challenging computer vision problem. A recent deep-based approach utilized matrix equivariant architectures for simultaneous recovery of camera pose and 3D scene structure from large image collections. That work, however, made the unrealistic assumption that the point tracks given as input are almost clean of outliers. Here, we propose an architecture suited to dealing with outliers by adding a multiview inlier/outlier classification module that respects the model equivariance and by utilizing a robust bundle adjustment step. Experiments demonstrate that our method can be applied successfully in realistic settings that include large image collections and point tracks extracted with common heuristics that include many outliers, achieving state-of-the-art accuracies in almost all runs, superior to existing deep-based methods and on-par with leading classical (non-deep) sequential and global methods.
|
[
{
"version": "v1",
"created": "Mon, 22 Apr 2024 15:29:19 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:35:20 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Khatib",
"Fadi",
""
],
[
"Kasten",
"Yoni",
""
],
[
"Moran",
"Dror",
""
],
[
"Galun",
"Meirav",
""
],
[
"Basri",
"Ronen",
""
]
] |
2404.14744
|
Carlos Silva-Acu\~na
|
Victoria Quir\'os-Cordero, Esteban Rojas-Gatjens, Mart\'in G\'omez-Dominguez, Hao Li, Carlo A. R. Perini, Natalie Stingelin, Juan-Pablo Correa-Baena, Eric R. Bittner, Ajay Ram Srimath Kandada and Carlos Silva-Acu\~na
|
Resolving exciton and polariton multi-particle correlations in an optical microcavity in the strong coupling regime
|
Main manuscript plus supplementary material document. Submitted for publication to Physical Review
| null | null | null |
cond-mat.mtrl-sci physics.chem-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Multi-particle correlations of exciton-polaritons and reservoir-excitons in the strong light-matter coupling regime dictate the quantum dynamics of optical microcavities. In this letter, we examine the many-body exciton-polariton dynamics in a Fabry-P\'erot microcavity of a two-dimensional metal-halide semiconductor over timescales involving polariton ($\ll 1$\,ps) and exciton ($\gg 1$\,ps) scattering. We find enhanced exciton nonlinear dynamics in the microcavity versus the bare semiconductor, concomitant with ultrafast polariton scattering dynamics. We measure, by means of coherent spectroscopy, the coupling between exciton-polaritons, bright excitons, and reservoir-excitons that highlight the complex scattering landscape that fundamentally drives polariton condensation.
|
[
{
"version": "v1",
"created": "Tue, 23 Apr 2024 04:51:33 GMT"
},
{
"version": "v2",
"created": "Tue, 16 Jul 2024 14:43:29 GMT"
},
{
"version": "v3",
"created": "Mon, 07 Apr 2025 02:47:25 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 13:53:02 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Quirós-Cordero",
"Victoria",
""
],
[
"Rojas-Gatjens",
"Esteban",
""
],
[
"Gómez-Dominguez",
"Martín",
""
],
[
"Li",
"Hao",
""
],
[
"Perini",
"Carlo A. R.",
""
],
[
"Stingelin",
"Natalie",
""
],
[
"Correa-Baena",
"Juan-Pablo",
""
],
[
"Bittner",
"Eric R.",
""
],
[
"Kandada",
"Ajay Ram Srimath",
""
],
[
"Silva-Acuña",
"Carlos",
""
]
] |
2404.17016
|
Gereon Ko{\ss}mann
|
Gereon Ko{\ss}mann and Ren\'e Schwonnek
|
Optimising the relative entropy under semidefinite constraints
|
generalization to f-divergences and corrected typos
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Finding the minimal relative entropy of two quantum states under semidefinite constraints is a pivotal problem located at the mathematical core of various applications in quantum information theory. An efficient method for providing provable upper and lower bounds is the central result of this work. Our primordial motivation stems from the essential task of estimating secret key rates for QKD from the measurement statistics of a real device. Further applications include the computation of channel capacities, the estimation of entanglement measures and many more. We build on a recently introduced integral representation of quantum relative entropy by [Frenkel, Quantum 7, 1102 (2023)] and provide reliable bounds as a sequence of semidefinite programs (SDPs). Our approach ensures provable sublinear convergence in the discretization, while also maintaining resource efficiency in terms of SDP matrix dimensions. Additionally, we can provide gap estimates to the optimum at each iteration stage.
|
[
{
"version": "v1",
"created": "Thu, 25 Apr 2024 20:19:47 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 10:12:30 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Koßmann",
"Gereon",
""
],
[
"Schwonnek",
"René",
""
]
] |
2404.17413
|
Federico Fioravanti
|
Ulle Endriss, Federico Fioravanti
|
Voting with Partial Orders: The Plurality and Anti-Plurality Classes
| null | null | null | null |
cs.GT econ.TH
|
http://creativecommons.org/licenses/by/4.0/
|
In the theory of voting, the Plurality rule for preferences that come in the form of linear orders selects the alternatives most frequently appearing in the first position of those orders, while the Anti-Plurality rule selects the alternatives least often occurring in the final position. We explore extensions of these rules to preferences that are partial orders, offering axiomatic characterisations for them.
|
[
{
"version": "v1",
"created": "Fri, 26 Apr 2024 13:43:26 GMT"
},
{
"version": "v2",
"created": "Mon, 29 Apr 2024 08:24:59 GMT"
},
{
"version": "v3",
"created": "Tue, 04 Jun 2024 10:48:32 GMT"
},
{
"version": "v4",
"created": "Tue, 07 Jan 2025 18:12:54 GMT"
},
{
"version": "v5",
"created": "Thu, 21 Aug 2025 10:03:48 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Endriss",
"Ulle",
""
],
[
"Fioravanti",
"Federico",
""
]
] |
2404.18021
|
Kaixuan Huang
|
Yuanhao Qu, Kaixuan Huang, Ming Yin, Kanghong Zhan, Dyllan Liu, Di Yin, Henry C. Cousins, William A. Johnson, Xiaotong Wang, Mihir Shah, Russ B. Altman, Denny Zhou, Mengdi Wang, Le Cong
|
CRISPR-GPT for Agentic Automation of Gene-editing Experiments
|
Accepted to Nature Biomedical Engineering
| null | null | null |
cs.AI cs.CL cs.HC q-bio.QM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The introduction of genome engineering technology has transformed biomedical research, making it possible to make precise changes to genetic information. However, creating an efficient gene-editing system requires a deep understanding of CRISPR technology, and the complex experimental systems under investigation. While Large Language Models (LLMs) have shown promise in various tasks, they often lack specific knowledge and struggle to accurately solve biological design problems. In this work, we introduce CRISPR-GPT, an LLM agent augmented with domain knowledge and external tools to automate and enhance the design process of CRISPR-based gene-editing experiments. CRISPR-GPT leverages the reasoning ability of LLMs to facilitate the process of selecting CRISPR systems, designing guide RNAs, recommending cellular delivery methods, drafting protocols, and designing validation experiments to confirm editing outcomes. We showcase the potential of CRISPR-GPT for assisting non-expert researchers with gene-editing experiments from scratch and validate the agent's effectiveness in a real-world use case. Furthermore, we explore the ethical and regulatory considerations associated with automated gene-editing design, highlighting the need for responsible and transparent use of these tools. Our work aims to bridge the gap between beginner biological researchers and CRISPR genome engineering techniques, and demonstrate the potential of LLM agents in facilitating complex biological discovery tasks. The published version of this draft is available at https://www.nature.com/articles/s41551-025-01463-z.
|
[
{
"version": "v1",
"created": "Sat, 27 Apr 2024 22:59:17 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 04:15:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Qu",
"Yuanhao",
""
],
[
"Huang",
"Kaixuan",
""
],
[
"Yin",
"Ming",
""
],
[
"Zhan",
"Kanghong",
""
],
[
"Liu",
"Dyllan",
""
],
[
"Yin",
"Di",
""
],
[
"Cousins",
"Henry C.",
""
],
[
"Johnson",
"William A.",
""
],
[
"Wang",
"Xiaotong",
""
],
[
"Shah",
"Mihir",
""
],
[
"Altman",
"Russ B.",
""
],
[
"Zhou",
"Denny",
""
],
[
"Wang",
"Mengdi",
""
],
[
"Cong",
"Le",
""
]
] |
2404.18972
|
Yasemin Vardar
|
Jan D. A. Vuik, Daan M. Pool and Y. Vardar
|
Impact of whole-body vibrations on electrovibration perception varies with target stimulus duration
|
28 pages; 7 figures, journal
| null |
10.1177/00187208251326662
| null |
cs.HC cs.RO cs.SY eess.SY
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This study explores the impact of whole-body vibrations induced by external vehicle perturbations, such as aircraft turbulence, on the perception of electrovibration displayed on touchscreens. Electrovibration holds promise as a technology for providing tactile feedback on future touchscreens, addressing usability challenges in vehicle cockpits. However, its performance under dynamic conditions, such as during whole-body vibrations induced by turbulence, still needs to be explored. We measured the absolute detection thresholds of 15 human participants for short- and long-duration electrovibration stimuli displayed on a touchscreen, both in the absence and presence of two types of turbulence motion generated by a motion simulator. Concurrently, we measured participants' applied contact force and finger scan speeds. Significantly higher (38%) absolute detection thresholds were observed for short electrovibration stimuli than for long stimuli. Finger scan speeds in the direction of turbulence, applied forces, and force fluctuation rates increased during whole-body vibrations due to biodynamic feedthrough. As a result, turbulence also significantly increased the perception thresholds, but only for short-duration electrovibration stimuli. The results reveal that whole-body vibrations can impede the perception of short-duration electrovibration stimuli, due to involuntary finger movements and increased normal force fluctuations. Our findings offer valuable insights for the future design of touchscreens with tactile feedback in vehicle cockpits.
|
[
{
"version": "v1",
"created": "Mon, 29 Apr 2024 14:09:11 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Vuik",
"Jan D. A.",
""
],
[
"Pool",
"Daan M.",
""
],
[
"Vardar",
"Y.",
""
]
] |
2404.19058
|
Funda Gultepe
|
Pritam Ghosh, Funda G\"ultepe
|
Geometry of extensions of free groups via automorphisms with fixed points on the complex of free factors
|
Final version accepted for publication. To appear in Algebraic and Geometric Topology
| null | null | null |
math.GR math.GT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We give conditions of an extension of a free group to be hyperbolic and relatively hyperbolic using the dynamics of the action of $\out$ on the complex of free factors combined with the weak attraction theory. We work with subgroups of exponentially growing outer automorphisms and instead of using a standard pingpong argument with loxodromics, we allow fixed points for the action and investigate the geometry of the extension group when the fixed points of the automorphisms on the complex of free factors are sufficiently far apart.
|
[
{
"version": "v1",
"created": "Mon, 29 Apr 2024 18:54:37 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 17:57:38 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ghosh",
"Pritam",
""
],
[
"Gültepe",
"Funda",
""
]
] |
2405.00826
|
Peter Bradshaw
|
Peter Bradshaw, Alexander Clow, Ladislav Stacho
|
A cornering strategy for synchronizing a DFA
|
24 pages
| null | null | null |
cs.FL math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper considers the existence of short synchronizing words in deterministic finite automata (DFAs). We define two general strategies for generating synchronizing words, and we show that each of these strategies can be applied if and only if a DFA is synchronizable. Furthermore, we show that if a synchronizable DFA is well-structured, then our strategies generate short synchronizing words. The first of our strategies, called the cornering strategy, takes advantage of states in a DFA with properties similar to those of a polytope vertex. The second of our strategies, similar to the cornering strategy and called the $f$-ordered strategy, takes advantage of a partial order defined on the states of a DFA.
We apply our cornering strategy to the class of difference DFAs, whose states form subsets of $\mathbb R^d$ and whose input symbols correspond to translation vectors between states. We show that difference DFAs share many similarities with aperiodic DFAs, and in particular, a difference DFA $M$ has a synchronizing word if and only if it has a universally reachable state. Using the cornering strategy, we also show that under certain conditions, such an $n$-state DFA $M$ has a synchronizing word of length at most $(n-1)^2$ and thereby satisfies \v{C}ern\'y's conjecture. Using the $f$-ordered strategy, we also show that a synchronizable DFA whose states have a certain partial order that is preserved by a set of short words also has a short synchronizing word, and we consider several consequences of this result.
Finally, we consider how the cornering strategy can be applied to the problem of synchronizing the product of two DFAs $M_1, M_2$ that share a common alphabet, and we show that the product $M_1 \times M_2$ often has a synchronizing word that is subquadratic in the number of states of $M_1 \times M_2$.
|
[
{
"version": "v1",
"created": "Wed, 01 May 2024 19:14:51 GMT"
},
{
"version": "v2",
"created": "Wed, 21 May 2025 01:09:41 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 21:24:03 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Bradshaw",
"Peter",
""
],
[
"Clow",
"Alexander",
""
],
[
"Stacho",
"Ladislav",
""
]
] |
2405.01810
|
Tian Xie
|
Tian Xie, Xueru Zhang
|
Non-linear Welfare-Aware Strategic Learning
|
Minor changes
| null | null | null |
cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
This paper studies algorithmic decision-making in the presence of strategic individual behaviors, where an ML model is used to make decisions about human agents and the latter can adapt their behavior strategically to improve their future data. Existing results on strategic learning have largely focused on the linear setting where agents with linear labeling functions best respond to a (noisy) linear decision policy. Instead, this work focuses on general non-linear settings where agents respond to the decision policy with only "local information" of the policy. Moreover, we simultaneously consider the objectives of maximizing decision-maker welfare (model prediction accuracy), social welfare (agent improvement caused by strategic behaviors), and agent welfare (the extent that ML underestimates the agents). We first generalize the agent best response model in previous works to the non-linear setting, then reveal the compatibility of welfare objectives. We show the three welfare can attain the optimum simultaneously only under restrictive conditions which are challenging to achieve in non-linear settings. The theoretical results imply that existing works solely maximizing the welfare of a subset of parties inevitably diminish the welfare of the others. We thus claim the necessity of balancing the welfare of each party in non-linear settings and propose an irreducible optimization algorithm suitable for general strategic learning. Experiments on synthetic and real data validate the proposed algorithm.
|
[
{
"version": "v1",
"created": "Fri, 03 May 2024 01:50:03 GMT"
},
{
"version": "v2",
"created": "Tue, 13 Aug 2024 19:19:07 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 13:20:26 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Xie",
"Tian",
""
],
[
"Zhang",
"Xueru",
""
]
] |
2405.01982
|
Xiangsheng Xu
|
Xiangsheng Xu
|
Global regularity for solutions of magnetohydrodynamic equations with large initial data
|
arXiv admin note: substantial text overlap with arXiv:2404.16433
| null | null | null |
math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the regularity properties of solutions to the initial value problem for the magnetohydrodynamic equations in $\mathbb{R}^N, N\geq 3$. We obtain a global in-time strong solution without any smallness assumptions on the initial data.
|
[
{
"version": "v1",
"created": "Fri, 03 May 2024 10:19:17 GMT"
},
{
"version": "v2",
"created": "Mon, 06 May 2024 23:09:44 GMT"
},
{
"version": "v3",
"created": "Thu, 09 May 2024 07:34:26 GMT"
},
{
"version": "v4",
"created": "Mon, 15 Jul 2024 17:10:13 GMT"
},
{
"version": "v5",
"created": "Sat, 02 Nov 2024 14:20:06 GMT"
},
{
"version": "v6",
"created": "Thu, 21 Aug 2025 15:49:52 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Xu",
"Xiangsheng",
""
]
] |
2405.03603
|
Yi Zhou
|
Yi Zhou, Taojun Hu, Yuji Sakamoto, Ao Huang, Xiao-Hua Zhou, Satoshi Hattori
|
Copas-Heckman-type sensitivity analysis for publication bias in rare-event meta-analysis under generalized linear mixed models
| null | null | null | null |
stat.ME stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
In systematic reviews and meta-analyses, publication bias (PB) is one of the serious concerns and mainly induced by selective publication of academic literatures. Although many methods have been proposed to deal with PB, almost all the methods are based on the normal-normal (NN) random-effects model assuming that data are normally distributed in both the within-study and the between-study levels. For rare-event meta-analysis where data contain rare occurrences of events, the standard NN random-effects model may perform poorly. Instead, some generalized linear mixed models (GLMMs) which employ the exact distribution for the number of events in within-study level provide alternatives and have been widely used in practice. However, limited methods can be applied to deal with PB in the GLMMs. To address this limitation, we propose a framework of sensitivity analysis for evaluating the impact of PB in various GLMMs. The proposed framework is developed based on the famous Copas-Heckman-type sensitivity analysis methods and can be easily implemented with the standard software with small computational cost. In this paper, we conduct simulation studies to assess the performance of proposed methods in adjusting PB and compare the results with related existing methods. Several real-world examples are also analyzed to show the broad applicability of our proposal in evaluating the potential impact of PB in meta-analysis of odds ratios and proportions with rare-event outcomes.
|
[
{
"version": "v1",
"created": "Mon, 06 May 2024 16:16:57 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 04:25:32 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhou",
"Yi",
""
],
[
"Hu",
"Taojun",
""
],
[
"Sakamoto",
"Yuji",
""
],
[
"Huang",
"Ao",
""
],
[
"Zhou",
"Xiao-Hua",
""
],
[
"Hattori",
"Satoshi",
""
]
] |
2405.04088
|
Anastasia Doikou
|
Anastasia Doikou
|
Parametric set-theoretic Yang-Baxter equation: p-racks, solutions & quantum algebras
|
33 pages, LaTex. Clarifications added, typos corrected
| null | null | null |
math-ph math.MP math.QA math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The theory of the parametric set-theoretic Yang-Baxter equation is established from a purely algebraic point of view. The first step towards this objective is the introduction of certain generalizations of the familiar shelves and racks called parametric ($p$)-shelves and racks. These objects satisfy a parametric self-distributivity condition and lead to solutions of the Yang-Baxter equation. Novel, non-reversible solutions are obtained from p-shelf/rack solutions by a suitable parametric twist, whereas all reversible set-theoretic solutions are reduced to the identity map via a parametric twist. The universal algebras associated to both $p$-rack and generic parametric, set-theoretic solutions are next presented and the corresponding universal R-matrices are derived. The admissible universal Drinfel'd twist is constructed allowing the derivation of the general set-theoretic universal R-matrix. By introducing the concept of a parametric coproduct we prove the existence of a parametric co-associativity. We show that the parametric coproduct is an algebra homomorphism and the universal R-matrices satisfy intertwining relations with the algebra coproducts.
|
[
{
"version": "v1",
"created": "Tue, 07 May 2024 07:42:59 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:23:30 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Doikou",
"Anastasia",
""
]
] |
2405.04244
|
Carles Roch I Carceller
|
Carles Roch I Carceller, Lucas Nunes Faria, Zheng-Hao Liu, Nicol\`o Sguerso, Ulrik Lund Andersen, Jonas Schou Neergaard-Nielsen and Jonatan Bohr Brask
|
Improving semi-device-independent randomness certification by entropy accumulation
|
18 pages, 5 figures, comments are welcome
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Certified randomness guaranteed to be unpredictable by adversaries is central to information security. The fundamental randomness inherent in quantum physics makes certification possible from devices that are only weakly characterised, i.e. requiring little trust in their implementation. It was recently shown that the amount of certifiable randomness can be greatly improved using the so-called Entropy Accumulation Theorem generalised to prepare-and-measure settings. Furthermore, this approach allows a finite-size analysis which avoids assuming that all rounds are independent and identically distributed. Here, we demonstrate this improvement in semi-device-independent randomness certification from untrusted measurements.
|
[
{
"version": "v1",
"created": "Tue, 07 May 2024 12:00:22 GMT"
},
{
"version": "v2",
"created": "Tue, 05 Nov 2024 12:58:05 GMT"
},
{
"version": "v3",
"created": "Thu, 01 May 2025 09:13:55 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 12:12:44 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Carceller",
"Carles Roch I",
""
],
[
"Faria",
"Lucas Nunes",
""
],
[
"Liu",
"Zheng-Hao",
""
],
[
"Sguerso",
"Nicolò",
""
],
[
"Andersen",
"Ulrik Lund",
""
],
[
"Neergaard-Nielsen",
"Jonas Schou",
""
],
[
"Brask",
"Jonatan Bohr",
""
]
] |
2405.08656
|
Boqian Zhang
|
Boqian Zhang and Juanmian Lei
|
Interpretable data-driven turbulence modeling for separated flows using symbolic regression with unit constraints
| null | null |
10.1007/s00707-025-04325-6
| null |
physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Machine learning techniques have been applied to enhance turbulence modeling in recent years. However, the "black box" nature of most machine learning techniques poses significant interpretability challenges in improving turbulence models. This paper introduces a novel unit-constrained turbulence modeling framework using symbolic regression to overcome these challenges. The framework amends the constitutive equation of linear eddy viscosity models (LEVMs) by establishing explicit equations between the Reynolds stress deviation and mean flow quantities, thereby improving the LEVM model's predictive capability for large separated turbulence. Unit consistency constraints are applied to the symbolic expressions to ensure physical realizability. The effectiveness of the framework and the generalization capability of the learned model are demonstrated through its application to the separated flow over 2D periodic hills and a backward-facing step. Compared to the standard k-{\epsilon} model, the learned model shows significantly improved predictive accuracy for anisotropic Reynolds stresses, velocity and skin friction, while exhibiting promising generalization capabilities across various scenarios.
|
[
{
"version": "v1",
"created": "Tue, 14 May 2024 14:35:12 GMT"
},
{
"version": "v2",
"created": "Sat, 25 May 2024 11:19:32 GMT"
},
{
"version": "v3",
"created": "Thu, 10 Oct 2024 12:27:02 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 10:00:12 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhang",
"Boqian",
""
],
[
"Lei",
"Juanmian",
""
]
] |
2405.13073
|
Edward Hall\'e-Hannan
|
Edward Hall\'e-Hannan, Charles Audet, Youssef Diouane, S\'ebastien Le Digabel, Paul Saves
|
A distance for mixed-variable and hierarchical domains with meta variables
|
29 pages (without references), 12 figures, 5 tables, data and scripts available at https://github.com/bbopt/graph_distance
|
Neurocomputing, Volume 653, 2025, 131208
|
10.1016/j.neucom.2025.131208
|
131208
|
stat.ML cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Heterogeneous datasets emerge in various machine learning and optimization applications that feature different input sources, types or formats. Most models or methods do not natively tackle heterogeneity. Hence, such datasets are often partitioned into smaller and simpler ones, which may limit the generalizability or performance, especially when data is limited. The first main contribution of this work is a modeling framework that generalizes hierarchical, tree-structured, variable-size or conditional search frameworks. The framework models mixed-variable and hierarchical domains in which variables may be continuous, integer, or categorical, with some identified as meta when they influence the structure of the problem. The second main contribution is a novel distance that compares any pair of mixed-variable points that do not share the same variables, allowing to use whole heterogeneous datasets that reside in mixed-variable and hierarchical domains with meta variables. The contributions are illustrated through regression and classification experiments using simple distance-based models applied to datasets of hyperparameters with corresponding performance scores.
|
[
{
"version": "v1",
"created": "Mon, 20 May 2024 23:11:03 GMT"
},
{
"version": "v2",
"created": "Mon, 19 Aug 2024 20:04:32 GMT"
},
{
"version": "v3",
"created": "Mon, 31 Mar 2025 15:41:59 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hallé-Hannan",
"Edward",
""
],
[
"Audet",
"Charles",
""
],
[
"Diouane",
"Youssef",
""
],
[
"Digabel",
"Sébastien Le",
""
],
[
"Saves",
"Paul",
""
]
] |
2405.13400
|
S\'andor Baran
|
Martin Leutbecher and S\'andor Baran
|
Ensemble size dependence of the logarithmic score for forecasts issued as multivariate normal distributions
|
24 pages; 7 figures; 4 tables
|
Quarterly Journal of the Royal Meteorological Society 151 (2025), no. 766, paper e4898
|
10.1002/qj.4898
| null |
stat.ME stat.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Multivariate probabilistic verification is concerned with the evaluation of joint probability distributions of vector quantities such as a weather variable at multiple locations or a wind vector for instance. The logarithmic score is a proper score that is useful in this context. In order to apply this score to ensemble forecasts, a choice for the density is required. Here, we are interested in the specific case when the density is multivariate normal with mean and covariance given by the ensemble mean and ensemble covariance, respectively. Under the assumptions of multivariate normality and exchangeability of the ensemble members, a relationship is derived which describes how the logarithmic score depends on ensemble size. It permits to estimate the score in the limit of infinite ensemble size from a small ensemble and thus produces a fair logarithmic score for multivariate ensemble forecasts under the assumption of normality. This generalises a study from 2018 which derived the ensemble size adjustment of the logarithmic score in the univariate case.
An application to medium-range forecasts examines the usefulness of the ensemble size adjustments when multivariate normality is only an approximation. Predictions of vectors consisting of several different combinations of upper air variables are considered. Logarithmic scores are calculated for these vectors using ECMWF's daily extended-range forecasts which consist of a 100-member ensemble. The probabilistic forecasts of these vectors are verified against operational ECMWF analyses in the Northern mid-latitudes in autumn 2023. Scores are computed for ensemble sizes from 8 to 100. The fair logarithmic scores of ensembles with different cardinalities are very close, in contrast to the unadjusted scores which decrease considerably with ensemble size. This provides evidence for the practical usefulness of the derived relationships.
|
[
{
"version": "v1",
"created": "Wed, 22 May 2024 07:19:18 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Leutbecher",
"Martin",
""
],
[
"Baran",
"Sándor",
""
]
] |
2405.14802
|
Hongxu Jiang
|
Hongxu Jiang, Muhammad Imran, Teng Zhang, Yuyin Zhou, Muxuan Liang, Kuang Gong, Wei Shao
|
Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image Generation
| null | null | null | null |
eess.IV cs.CV
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Denoising diffusion probabilistic models (DDPMs) have achieved unprecedented success in computer vision. However, they remain underutilized in medical imaging, a field crucial for disease diagnosis and treatment planning. This is primarily due to the high computational cost associated with (1) the use of large number of time steps (e.g., 1,000) in diffusion processes and (2) the increased dimensionality of medical images, which are often 3D or 4D. Training a diffusion model on medical images typically takes days to weeks, while sampling each image volume takes minutes to hours. To address this challenge, we introduce Fast-DDPM, a simple yet effective approach capable of improving training speed, sampling speed, and generation quality simultaneously. Unlike DDPM, which trains the image denoiser across 1,000 time steps, Fast-DDPM trains and samples using only 10 time steps. The key to our method lies in aligning the training and sampling procedures to optimize time-step utilization. Specifically, we introduced two efficient noise schedulers with 10 time steps: one with uniform time step sampling and another with non-uniform sampling. We evaluated Fast-DDPM across three medical image-to-image generation tasks: multi-image super-resolution, image denoising, and image-to-image translation. Fast-DDPM outperformed DDPM and current state-of-the-art methods based on convolutional networks and generative adversarial networks in all tasks. Additionally, Fast-DDPM reduced the training time to 0.2x and the sampling time to 0.01x compared to DDPM. Our code is publicly available at: https://github.com/mirthAI/Fast-DDPM.
|
[
{
"version": "v1",
"created": "Thu, 23 May 2024 17:12:22 GMT"
},
{
"version": "v2",
"created": "Fri, 24 May 2024 02:48:33 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 14:56:13 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jiang",
"Hongxu",
""
],
[
"Imran",
"Muhammad",
""
],
[
"Zhang",
"Teng",
""
],
[
"Zhou",
"Yuyin",
""
],
[
"Liang",
"Muxuan",
""
],
[
"Gong",
"Kuang",
""
],
[
"Shao",
"Wei",
""
]
] |
2405.18168
|
Philippos Papaphilippou
|
Philippos Papaphilippou, Wayne Luk
|
Efficient Adaptable High-throughput Streaming Aggregation Engines
| null | null | null | null |
cs.AR cs.DB
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Aggregation queries are a series of computationally-demanding analytics operations on counted, grouped or time series data. They include tasks such as summation or finding the median among the items of the same group, and within a specified number of the last observed tuples for sliding window aggregation (SWAG). They have a wide range of applications including database analytics, operating systems, bank security and medical sensors. Existing challenges include the hardware complexity that comes with efficiently handling per-group states using hash-based approaches. This paper presents an adaptable pipeline for calculating a wide range of aggregation queries with high throughput. It is then adapted for SWAG and achieves up to 476x speedup over the CPU core of the same platform. It achieves unparalleled levels of performance and functionality such as a throughput of 1 GT/s on our setup for SWAG without groups, and more advanced operators with up to 4x the window sizes than the state-of-the-art with groups as an approximation for SWAG featuring per-group windows using a fraction of the resources and no DRAM.
|
[
{
"version": "v1",
"created": "Tue, 28 May 2024 13:29:31 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Nov 2024 06:16:27 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 11:19:44 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Papaphilippou",
"Philippos",
""
],
[
"Luk",
"Wayne",
""
]
] |
2406.00443
|
Nuno Fachada
|
Nuno Fachada, Ant\'onio R. Rodrigues, Diogo de Andrade, Phil Lopes
|
Generating 3D Terrain with 2D Cellular Automata
|
The peer-reviewed version of this paper is published in IEEE Xplore at https://doi.org/10.1109/CoG64752.2025.11114361. This version is typeset by the author and differs only in pagination and typographical detail
|
2025 IEEE Conference on Games (CoG), 1-4, IEEE, 2025
|
10.1109/CoG64752.2025.11114361
| null |
nlin.CG cs.AI cs.GR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper explores the use of 2D cellular automata (CA) to generate 3D terrains through a simple additive approach. Experimenting with multiple CA transition rules produced aesthetically interesting, navigable landscapes, suggesting applicability for terrain generation in games.
|
[
{
"version": "v1",
"created": "Sat, 01 Jun 2024 13:43:28 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:01:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fachada",
"Nuno",
""
],
[
"Rodrigues",
"António R.",
""
],
[
"de Andrade",
"Diogo",
""
],
[
"Lopes",
"Phil",
""
]
] |
2406.02394
|
Maxime Griot
|
Maxime Griot, Jean Vanderdonckt, Demet Yuksel, Coralie Hemptinne
|
Pattern Recognition or Medical Knowledge? The Problem with Multiple-Choice Questions in Medicine
|
ACL 2025 main
| null |
10.18653/v1/2025.acl-long.266
| null |
cs.CL cs.AI cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Large Language Models (LLMs) such as ChatGPT demonstrate significant potential in the medical domain and are often evaluated using multiple-choice questions (MCQs) modeled on exams like the USMLE. However, such benchmarks may overestimate true clinical understanding by rewarding pattern recognition and test-taking heuristics. To investigate this, we created a fictional medical benchmark centered on an imaginary organ, the Glianorex, allowing us to separate memorized knowledge from reasoning ability. We generated textbooks and MCQs in English and French using leading LLMs, then evaluated proprietary, open-source, and domain-specific models in a zero-shot setting. Despite the fictional content, models achieved an average score of 64%, while physicians scored only 27%. Fine-tuned medical models outperformed base models in English but not in French. Ablation and interpretability analyses revealed that models frequently relied on shallow cues, test-taking strategies, and hallucinated reasoning to identify the correct choice. These results suggest that standard MCQ-based evaluations may not effectively measure clinical reasoning and highlight the need for more robust, clinically meaningful assessment methods for LLMs.
|
[
{
"version": "v1",
"created": "Tue, 04 Jun 2024 15:08:56 GMT"
},
{
"version": "v2",
"created": "Sat, 31 May 2025 11:21:22 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Griot",
"Maxime",
""
],
[
"Vanderdonckt",
"Jean",
""
],
[
"Yuksel",
"Demet",
""
],
[
"Hemptinne",
"Coralie",
""
]
] |
2406.06096
|
Wenbo Lin
|
Wenbo Lin, Nilo Mata-Cervera, Yasutomo Ota, Yijie Shen, Satoshi Iwamoto
|
Space-Time Hopfion Crystals
|
6 pages, 4 figures
| null |
10.1103/hh5s-cprt
| null |
physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
Hopfions, higher-dimensional topological quasiparticles with sophisticated 3D knotted spin textures discovered in condensed matter and photonic systems, show promise in high-density data storage and transfer. Here we present crystalline structures of hopfions lying in space-time constructed by spatiotemporally structured light. A practical methodology using bichromatic structured light beams or dipole arrays to assemble 1D and higher dimensional hopfion lattices is proposed and a technique for tailoring topological orders is elucidated. The birth of photonic hopfion crystals heralds a new era in high-dimensional, condensed, and robust topological information processing.
|
[
{
"version": "v1",
"created": "Mon, 10 Jun 2024 08:27:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Lin",
"Wenbo",
""
],
[
"Mata-Cervera",
"Nilo",
""
],
[
"Ota",
"Yasutomo",
""
],
[
"Shen",
"Yijie",
""
],
[
"Iwamoto",
"Satoshi",
""
]
] |
2406.08017
|
Mikhail Borovoi
|
Mikhail Borovoi and Jean-Louis Colliot-Th\'el\`ene
|
The defect of weak approximation for a reductive group over a global field
|
v7: 27 pages, including the listing of a Magma program; v8: 29 pages, including an appendix by J.-L. Colliot-Th\'el\`ene
| null | null |
MPIM-Bonn-2024
|
math.RT math.AG math.GR math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We compute the defect of weak approximation for a reductive group G over a global field K in terms of the algebraic fundamental group of G.
|
[
{
"version": "v1",
"created": "Wed, 12 Jun 2024 09:17:59 GMT"
},
{
"version": "v2",
"created": "Sat, 15 Jun 2024 18:54:32 GMT"
},
{
"version": "v3",
"created": "Sat, 06 Jul 2024 13:42:46 GMT"
},
{
"version": "v4",
"created": "Sun, 01 Sep 2024 14:59:45 GMT"
},
{
"version": "v5",
"created": "Wed, 04 Sep 2024 05:45:26 GMT"
},
{
"version": "v6",
"created": "Tue, 06 May 2025 10:38:23 GMT"
},
{
"version": "v7",
"created": "Wed, 13 Aug 2025 10:07:24 GMT"
},
{
"version": "v8",
"created": "Thu, 21 Aug 2025 12:55:44 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Borovoi",
"Mikhail",
""
],
[
"Colliot-Thélène",
"Jean-Louis",
""
]
] |
2406.08118
|
Junming Zhang
|
Junming Zhang
|
Non-maximal Anosov representations from surface groups to $\mathrm{SO}_0(2,3)$
|
Final version, to appear in Mathematische Annalen. 35 pages, comments are very welcome
| null | null | null |
math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove the representation given by a stable $\alpha_1$-cyclic parabolic $\mathrm{SO}_0(2,3)$-Higgs bundle through the non-Abelian Hodge correspondence is $\{\alpha_2\}$-almost dominated. This is a generalization of Filip's result on weight $3$ variation of Hodge structures and answers a question asked by Collier, Tholozan and Toulisse.
|
[
{
"version": "v1",
"created": "Wed, 12 Jun 2024 11:53:34 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Jun 2024 06:06:22 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 14:42:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhang",
"Junming",
""
]
] |
2406.09832
|
Hadrian Heine
|
Hadrian Heine
|
On bi-enriched $\infty$-categories
|
arXiv admin note: substantial text overlap with arXiv:2406.08925
| null | null | null |
math.CT math.AT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We extend Lurie's definition of enriched $\infty$-categories to notions of left enriched, right enriched and bienriched $\infty$-categories, which generalize the concepts of closed left tensored, right tensored and bitensored $\infty$-categories and share many desirable features with them. We use bienriched $\infty$-categories to endow the $\infty$-category of enriched functors with enrichment that generalizes both the internal hom of the tensor product of enriched $\infty$-categories when the latter exists, and the free cocompletion under colimits and tensors. As an application we construct enriched Kan-extensions from operadic Kan-extensions, compute the monad for enriched functors, prove an end formula for morphism objects of enriched $\infty$-categories of enriched functors and a coend formula for the relative tensor product of enriched profunctors and construct transfer of enrichment from scalar extension of presentably bitensored $\infty$-categories. In particular, we develop an independent theory of enriched $\infty$-categories for Lurie's model of enriched $\infty$-categories.
|
[
{
"version": "v1",
"created": "Fri, 14 Jun 2024 08:41:56 GMT"
},
{
"version": "v2",
"created": "Tue, 08 Oct 2024 22:32:21 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 17:40:29 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Heine",
"Hadrian",
""
]
] |
2406.10885
|
Weiqi Wang Mr.
|
Weiqi Wang, Tianqing Fang, Haochen Shi, Baixuan Xu, Wenxuan Ding, Liyu Zhang, Wei Fan, Jiaxin Bai, Haoran Li, Xin Liu, Yangqiu Song
|
On the Role of Entity and Event Level Conceptualization in Generalizable Reasoning: A Survey of Tasks, Methods, Applications, and Future Directions
|
Findings of EMNLP 2025
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Conceptualization, a fundamental element of human cognition, plays a pivotal role in human generalizable reasoning. Generally speaking, it refers to the process of sequentially abstracting specific instances into higher-level concepts and then forming abstract knowledge that can be applied in unfamiliar or novel situations. This enhances models' inferential capabilities and supports the effective transfer of knowledge across various domains. Despite its significance, the broad nature of this term has led to inconsistencies in understanding conceptualization across various works, as there exists different types of instances that can be abstracted in a wide variety of ways. There is also a lack of a systematic overview that comprehensively examines existing works on the definition, execution, and application of conceptualization to enhance reasoning tasks. In this paper, we address these gaps by first proposing a categorization of different types of conceptualizations into four levels based on the types of instances being conceptualized, in order to clarify the term and define the scope of our work. Then, we present the first comprehensive survey of over 150 papers, surveying various definitions, resources, methods, and downstream applications related to conceptualization into a unified taxonomy, with a focus on the entity and event levels. Furthermore, we shed light on potential future directions in this field and hope to garner more attention from the community.
|
[
{
"version": "v1",
"created": "Sun, 16 Jun 2024 10:32:41 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 23:24:04 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Weiqi",
""
],
[
"Fang",
"Tianqing",
""
],
[
"Shi",
"Haochen",
""
],
[
"Xu",
"Baixuan",
""
],
[
"Ding",
"Wenxuan",
""
],
[
"Zhang",
"Liyu",
""
],
[
"Fan",
"Wei",
""
],
[
"Bai",
"Jiaxin",
""
],
[
"Li",
"Haoran",
""
],
[
"Liu",
"Xin",
""
],
[
"Song",
"Yangqiu",
""
]
] |
2406.13240
|
Soichiro Fujii
|
Soichiro Fujii, Keisuke Hoshino, Yuki Maehara
|
$\omega$-weak equivalences between weak $\omega$-categories
|
33 pages. Major revision. Comments welcome!
|
Advances in Mathematics, 480:110490, 2025
|
10.1016/j.aim.2025.110490
| null |
math.CT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study $\omega$-weak equivalences between weak $\omega$-categories in the sense of Batanin-Leinster. Our $\omega$-weak equivalences are strict $\omega$-functors satisfying essential surjectivity in every dimension, and when restricted to those between strict $\omega$-categories, they coincide with the weak equivalences in the model category of strict $\omega$-categories defined by Lafont, M\'etayer, and Worytkiewicz. We show that the class of $\omega$-weak equivalences has the 2-out-of-3 property. We also consider a generalisation of $\omega$-weak equivalences, defined as weak $\omega$-functors (in the sense of Garner) satisfying essential surjectivity, and show that this class also has the 2-out-of-3 property.
|
[
{
"version": "v1",
"created": "Wed, 19 Jun 2024 06:06:04 GMT"
},
{
"version": "v2",
"created": "Sat, 24 Aug 2024 01:44:50 GMT"
},
{
"version": "v3",
"created": "Mon, 02 Jun 2025 08:50:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fujii",
"Soichiro",
""
],
[
"Hoshino",
"Keisuke",
""
],
[
"Maehara",
"Yuki",
""
]
] |
2406.14159
|
S\'andor Baran
|
M\'aria Lakatos and S\'andor Baran
|
Enhancing multivariate post-processed visibility predictions utilizing CAMS forecasts
|
23 pages, 10 figures
|
Meteorological Applications 31 (2024), no. 6, paper e70015
|
10.1002/met.70015
| null |
stat.AP stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In our contemporary era, meteorological weather forecasts increasingly incorporate ensemble predictions of visibility - a parameter of great importance in aviation, maritime navigation, and air quality assessment, with direct implications for public health. However, this weather variable falls short of the predictive accuracy achieved for other quantities issued by meteorological centers. Therefore, statistical post-processing is recommended to enhance the reliability and accuracy of predictions. By estimating the predictive distributions of the variables with the aid of historical observations and forecasts, one can achieve statistical consistency between true observations and ensemble predictions. Visibility observations, following the recommendation of the World Meteorological Organization, are typically reported in discrete values; hence, the predictive distribution of the weather quantity takes the form of a discrete parametric law. Recent studies demonstrated that the application of classification algorithms can successfully improve the skill of such discrete forecasts; however, a frequently emerging issue is that certain spatial and/or temporal dependencies could be lost between marginals. Based on visibility ensemble forecasts of the European Centre for Medium-Range Weather Forecasts for 30 locations in Central Europe, we investigate whether the inclusion of Copernicus Atmosphere Monitoring Service (CAMS) predictions of the same weather quantity as an additional covariate could enhance the skill of the post-processing methods and whether it contributes to the successful integration of spatial dependence between marginals. Our study confirms that post-processed forecasts are substantially superior to raw and climatological predictions, and the utilization of CAMS forecasts provides a further significant enhancement both in the univariate and multivariate setup.
|
[
{
"version": "v1",
"created": "Thu, 20 Jun 2024 09:57:49 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Lakatos",
"Mária",
""
],
[
"Baran",
"Sándor",
""
]
] |
2406.17509
|
Nazife Ozdes Koca
|
Nazife Ozdes Koca and Mehmet Koca
|
Affine subgroups of the affine Coxeter group with the same Coxeter number
|
9 pages, 8 figures
|
Sultan Qaboos University Journal For Science (2025) Vol. 30: Iss. 1, 16-22
|
10.53539/2414-536X.1399
|
https://squjs.squ.edu.om/squjs/vol30/iss1/4
|
math-ph math.MP
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Affine subgroups having the same Coxeter number with the affine Coxeter groups W(An), W(Dn), and W(En) are constructed by graph folding technique. The affine groups W(Cn) and W(Bn) are obtained from the Coxeter groups W(A2n-1) and W(D2n-1) respectively. The affine groups W(E6), W(D6) and W(E8) lead to the affine groups W(F4), W(H3), and W(H4) respectively by graph folding. The latter two are the non-crystallographic groups where W(H3) plays a special role in the quasicrystallographic structures with icosahedral symmetry. A general construction of the affine dihedral subgroups is introduced, some of which, describe the existing planar quasicrystallography. In the construction of the root systems, sets of orthonormal vectors are used but a special non-orthogonal set of vectors in the formulation of the root system of W(An) is also introduced which has practical applications in the construction of the lattices An and An* and their Delone and Voronoi cells.
|
[
{
"version": "v1",
"created": "Tue, 25 Jun 2024 12:47:53 GMT"
},
{
"version": "v2",
"created": "Sun, 30 Jun 2024 12:00:03 GMT"
},
{
"version": "v3",
"created": "Sat, 20 Jul 2024 07:08:30 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Koca",
"Nazife Ozdes",
""
],
[
"Koca",
"Mehmet",
""
]
] |
2406.17840
|
Jiaman Li
|
Zhen Wu, Jiaman Li, Pei Xu, C. Karen Liu
|
Human-Object Interaction from Human-Level Instructions
|
ICCV 2025, project page: https://hoifhli.github.io/
| null | null | null |
cs.AI cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Intelligent agents must autonomously interact with the environments to perform daily tasks based on human-level instructions. They need a foundational understanding of the world to accurately interpret these instructions, along with precise low-level movement and interaction skills to execute the derived actions. In this work, we propose the first complete system for synthesizing physically plausible, long-horizon human-object interactions for object manipulation in contextual environments, driven by human-level instructions. We leverage large language models (LLMs) to interpret the input instructions into detailed execution plans. Unlike prior work, our system is capable of generating detailed finger-object interactions, in seamless coordination with full-body movements. We also train a policy to track generated motions in physics simulation via reinforcement learning (RL) to ensure physical plausibility of the motion. Our experiments demonstrate the effectiveness of our system in synthesizing realistic interactions with diverse objects in complex environments, highlighting its potential for real-world applications.
|
[
{
"version": "v1",
"created": "Tue, 25 Jun 2024 17:46:28 GMT"
},
{
"version": "v2",
"created": "Wed, 11 Dec 2024 04:37:15 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 08:23:55 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wu",
"Zhen",
""
],
[
"Li",
"Jiaman",
""
],
[
"Xu",
"Pei",
""
],
[
"Liu",
"C. Karen",
""
]
] |
2406.18456
|
Nan Wu
|
Pei-Cheng Kuo and Nan Wu
|
Boundary Detection Algorithm Inspired by Locally Linear Embedding
|
37 Pages, 10 figures
| null | null | null |
stat.ML math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In the study of high-dimensional data, it is often assumed that the data set possesses an underlying lower-dimensional structure. A practical model for this structure is an embedded compact manifold with boundary. Since the underlying manifold structure is typically unknown, identifying boundary points from the data distributed on the manifold is crucial for various applications. In this work, we propose a method for detecting boundary points inspired by the widely used locally linear embedding algorithm. We implement this method using two nearest neighborhood search schemes: the epsilon-radius ball scheme and the K-nearest neighbor scheme. This algorithm incorporates the geometric information of the data structure, particularly through its close relation with the local covariance matrix. We analyze the algorithm by exploring the spectral properties of the local covariance matrix, with the findings guiding the selection of key parameters. In the presence of high-dimensional noise, we propose a framework aimed at enhancing boundary detection in noisy data. Furthermore, we demonstrate the algorithm's performance with simulated examples.
|
[
{
"version": "v1",
"created": "Wed, 26 Jun 2024 16:05:57 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:01:46 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Kuo",
"Pei-Cheng",
""
],
[
"Wu",
"Nan",
""
]
] |
2407.02075
|
Pasquale De Marinis
|
Pasquale De Marinis, Nicola Fanelli, Raffaele Scaringi, Emanuele Colonna, Giuseppe Fiameni, Gennaro Vessio and Giovanna Castellano
|
Label Anything: Multi-Class Few-Shot Semantic Segmentation with Visual Prompts
|
ECAI 2025 - 28th European Conference on Artificial Intelligence
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Few-shot semantic segmentation aims to segment objects from previously unseen classes using only a limited number of labeled examples. In this paper, we introduce Label Anything, a novel transformer-based architecture designed for multi-prompt, multi-way few-shot semantic segmentation. Our approach leverages diverse visual prompts -- points, bounding boxes, and masks -- to create a highly flexible and generalizable framework that significantly reduces annotation burden while maintaining high accuracy. Label Anything makes three key contributions: ($\textit{i}$) we introduce a new task formulation that relaxes conventional few-shot segmentation constraints by supporting various types of prompts, multi-class classification, and enabling multiple prompts within a single image; ($\textit{ii}$) we propose a novel architecture based on transformers and attention mechanisms; and ($\textit{iii}$) we design a versatile training procedure allowing our model to operate seamlessly across different $N$-way $K$-shot and prompt-type configurations with a single trained model. Our extensive experimental evaluation on the widely used COCO-$20^i$ benchmark demonstrates that Label Anything achieves state-of-the-art performance among existing multi-way few-shot segmentation methods, while significantly outperforming leading single-class models when evaluated in multi-class settings. Code and trained models are available at https://github.com/pasqualedem/LabelAnything.
|
[
{
"version": "v1",
"created": "Tue, 02 Jul 2024 09:08:06 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Jul 2025 10:46:04 GMT"
},
{
"version": "v3",
"created": "Fri, 25 Jul 2025 13:01:46 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 14:39:31 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"De Marinis",
"Pasquale",
""
],
[
"Fanelli",
"Nicola",
""
],
[
"Scaringi",
"Raffaele",
""
],
[
"Colonna",
"Emanuele",
""
],
[
"Fiameni",
"Giuseppe",
""
],
[
"Vessio",
"Gennaro",
""
],
[
"Castellano",
"Giovanna",
""
]
] |
2407.03761
|
Vincenzo Reda
|
Marvin Anas Hahn and Vincenzo Reda
|
Universal piecewise polynomiality for counting curves in toric surfaces
|
24 pages, 12 figures, 3 tables
| null | null | null |
math.AG math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Inspired by piecewise polynomiality results of double Hurwitz numbers, Ardila and Brugall\'e introduced an enumerative problem which they call double Gromov--Witten invariants of Hirzebruch surfaces. These invariants serve as a two-dimensional analogue and satisfy a similar piecewise polynomial structure. More precisely, they introduced the enumeration of curves in Hirzebruch surfaces satisfying point conditions and tangency conditions on the two parallel toric boundaries. These conditions are stored in four partitions and the resulting invariants are piecewise polynomial in their entries. Moreover, they found that these expressions also behave polynomially with respect to the parameter determining the underlying Hirzebruch surfaces. Based on work of Ardila and Block, they proposed that such a polynomiality could also hold while changing between more general toric surfaces corresponding to $h$-transverse polygons. In this work, we answer this question affirmatively. Moreover, we express the resulting invariants for $h$-transverse polygons as matrix elements in the two-dimensional bosonic Fock space.
|
[
{
"version": "v1",
"created": "Thu, 04 Jul 2024 09:15:01 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 09:54:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hahn",
"Marvin Anas",
""
],
[
"Reda",
"Vincenzo",
""
]
] |
2407.04767
|
Mainak Mukhopadhyay
|
Mainak Mukhopadhyay, Shigeo S. Kimura, Brian D. Metzger
|
High-energy neutrino signatures from pulsar remnants of binary neutron-star mergers: coincident detection prospects with gravitational waves
|
23 pages, 8 figures; published version
|
Astrophys. J. 987 (2025) 2, 218
|
10.3847/1538-4357/adc913
| null |
astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Binary neutron-star (BNS) mergers are accompanied by multi-messenger emissions, including gravitational wave (GW), neutrino, and electromagnetic signals. Some fraction of BNS mergers may result in a rapidly spinning magnetar as a remnant, which can enhance both the EM and neutrino emissions. In this study, we model the neutrino emissions from such systems and discuss the prospects for detecting the neutrinos coincident with GW signatures. We consider a scenario where a magnetar remnant drives a pulsar wind using its spin energy. The wind interacts with the surrounding kilonova ejecta, forming a nebula filled with non-thermal photons. Ions and nuclei extracted from the magnetar's surface can be accelerated in the polar-cap and the termination-shock regions. We investigate the neutrino fluences resulting from photomeson interactions, where accelerated CR protons interact with the photons in the nebula. Our findings indicate that the peak neutrino fluence is $\sim 10^{-2}\rm GeV~cm^{-2}$ for a source at $40$ Mpc, which is reached approximately $\mathcal{O}( 1-10\ {\rm days})$ post merger. Finally, we examine the potential for GW-triggered stacking searches with IceCube-Gen2 using next-generation GW detectors such as the Cosmic Explorer (CE) and the Einstein Telescope (ET). We conclude that, assuming an optimistic neutrino emission model, a combination of CE+ET would offer a high probability of neutrino detection from these sources within an operational timescale of $\sim 20$ years. In case of non-detection, $2 \sigma$ level constraints on model parameters can be established within similar joint operation timescales.
|
[
{
"version": "v1",
"created": "Fri, 05 Jul 2024 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 19:44:01 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mukhopadhyay",
"Mainak",
""
],
[
"Kimura",
"Shigeo S.",
""
],
[
"Metzger",
"Brian D.",
""
]
] |
2407.04776
|
Ryan Steed
|
Ryan Steed, Diana Qing, Zhiwei Steven Wu
|
Quantifying Privacy Risks of Public Statistics to Residents of Subsidized Housing
|
Harvard Data Science Review, forthcoming
|
Harvard Data Science Review Special Issue 6 (August 2025)
|
10.1162/99608f92.39d8bfa4
| null |
cs.CY
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
As the U.S. Census Bureau implements its controversial new disclosure avoidance system, researchers and policymakers debate the necessity of new privacy protections for public statistics. With experiments on both public statistics and synthetic microdata, we explore a particular privacy concern: respondents in subsidized housing may deliberately not mention unauthorized children and other household members for fear of being discovered and evicted. By combining public statistics from the Decennial Census and the Department of Housing and Urban Development, we demonstrate a simple, inexpensive reconstruction attack that could identify subsidized households living in violation of occupancy guidelines in 2010. Experiments on synthetic data suggest that a random swapping mechanism similar to the Census Bureau's 2010 disclosure avoidance measures does not significantly reduce the precision of this attack, while a differentially private mechanism similar to the 2020 disclosure avoidance system does. Our results provide a valuable example for policymakers seeking trustworthy public statistics.
|
[
{
"version": "v1",
"created": "Fri, 05 Jul 2024 18:00:02 GMT"
},
{
"version": "v2",
"created": "Mon, 12 May 2025 16:01:29 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Steed",
"Ryan",
""
],
[
"Qing",
"Diana",
""
],
[
"Wu",
"Zhiwei Steven",
""
]
] |
2407.05383
|
You Wu
|
You Wu, Xucheng Wang, Dan Zeng, Hengzhou Ye, Xiaolan Xie, Qijun Zhao, and Shuiwang Li
|
Learning Motion Blur Robust Vision Transformers for Real-Time UAV Tracking
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Unmanned aerial vehicle (UAV) tracking is critical for applications like surveillance, search-and-rescue, and autonomous navigation. However, the high-speed movement of UAVs and targets introduces unique challenges, including real-time processing demands and severe motion blur, which degrade the performance of existing generic trackers. While single-stream vision transformer (ViT) architectures have shown promise in visual tracking, their computational inefficiency and lack of UAV-specific optimizations limit their practicality in this domain. In this paper, we boost the efficiency of this framework by tailoring it into an adaptive computation framework that dynamically exits Transformer blocks for real-time UAV tracking. The motivation behind this is that tracking tasks with fewer challenges can be adequately addressed using low-level feature representations. Simpler tasks can often be handled with less demanding, lower-level features. This approach allows the model use computational resources more efficiently by focusing on complex tasks and conserving resources for easier ones. Another significant enhancement introduced in this paper is the improved effectiveness of ViTs in handling motion blur, a common issue in UAV tracking caused by the fast movements of either the UAV, the tracked objects, or both. This is achieved by acquiring motion blur robust representations through enforcing invariance in the feature representation of the target with respect to simulated motion blur. We refer to our proposed approach as BDTrack. Extensive experiments conducted on four tracking benchmarks validate the effectiveness and versatility of our approach, demonstrating its potential as a practical and effective approach for real-time UAV tracking. Code is released at: https://github.com/wuyou3474/BDTrack.
|
[
{
"version": "v1",
"created": "Sun, 07 Jul 2024 14:10:04 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:44:52 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wu",
"You",
""
],
[
"Wang",
"Xucheng",
""
],
[
"Zeng",
"Dan",
""
],
[
"Ye",
"Hengzhou",
""
],
[
"Xie",
"Xiaolan",
""
],
[
"Zhao",
"Qijun",
""
],
[
"Li",
"Shuiwang",
""
]
] |
2407.06956
|
Tom de Jong
|
Tom de Jong and Mart\'in H\"otzel Escard\'o
|
Continuous and algebraic domains in univalent foundations
|
v4: Accepted for publication in the Journal of Pure and Applied Algebra. This revision adds an abridged version of arXiv:2407.06952 at the request of the journal
| null | null | null |
cs.LO math.LO
|
http://creativecommons.org/licenses/by/4.0/
|
We develop the theory of continuous and algebraic domains in constructive and predicative univalent foundations, building upon our earlier work on basic domain theory in this setting. That we work predicatively means that we do not assume Voevodsky's propositional resizing axioms. Our work is constructive in the sense that we do not rely on excluded middle or the axiom of (countable) choice. To deal with size issues and give a predicatively suitable definition of continuity of a dcpo, we follow Johnstone and Joyal's work on continuous categories. Adhering to the univalent perspective, we explicitly distinguish between data and property. To ensure that being continuous is a property of a dcpo, we turn to the propositional truncation, although we explain that some care is needed to avoid needing the axiom of choice. We also adapt the notion of a domain-theoretic basis to the predicative setting by imposing suitable smallness conditions, analogous to the categorical concept of an accessible category. All our running examples of continuous dcpos are then actually examples of dcpos with small bases which we show to be well behaved predicatively. In particular, such dcpos are exactly those presented by small ideals. As an application of the theory, we show that Scott's $D_\infty$ model of the untyped $\lambda$-calculus is an example of an algebraic dcpo with a small basis. Our work is formalised in the Agda proof assistant and its ability to infer universe levels has been invaluable for our purposes.
|
[
{
"version": "v1",
"created": "Tue, 09 Jul 2024 15:35:22 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Jul 2024 09:41:18 GMT"
},
{
"version": "v3",
"created": "Thu, 18 Jul 2024 11:46:43 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 15:54:38 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"de Jong",
"Tom",
""
],
[
"Escardó",
"Martín Hötzel",
""
]
] |
2407.07110
|
Yong-Yeon Jo Ph.D.
|
Junho Song, Jong-Hwan Jang, DongGyun Hong, Joon-myoung Kwon, and Yong-Yeon Jo
|
CREMA: A Contrastive Regularized Masked Autoencoder for Robust ECG Diagnostics across Clinical Domains
|
10 pages
| null | null | null |
cs.LG cs.AI eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Electrocardiogram (ECG) diagnosis remains challenging due to limited labeled data and the need to capture subtle yet clinically meaningful variations in rhythm and morphology. We present CREMA (Contrastive Regularized Masked Autoencoder), a foundation model for 12-lead ECGs designed to learn generalizable representations through self-supervised pretraining. CREMA combines generative learning and contrastive regularization via a Contrastive Regularized MAE loss, and employs a Signal Transformer (SiT) architecture to capture both local waveform details and global temporal dependencies. We evaluate CREMA on benchmark datasets and real-world clinical environments, including deployment scenarios with significant distribution shifts. CREMA outperforms supervised baselines and existing self-supervised models in both linear probing and fine-tuning evaluations. Notably, it maintains superior performance across diverse clinical domains, such as emergency care, highlighting its robustness under real-world conditions. These results demonstrate that CREMA serves as a scalable and reliable foundation model for ECG diagnostics, supporting downstream applications across heterogeneous and high-risk clinical settings.
|
[
{
"version": "v1",
"created": "Wed, 26 Jun 2024 02:24:13 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Oct 2024 09:33:39 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 13:24:18 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Song",
"Junho",
""
],
[
"Jang",
"Jong-Hwan",
""
],
[
"Hong",
"DongGyun",
""
],
[
"Kwon",
"Joon-myoung",
""
],
[
"Jo",
"Yong-Yeon",
""
]
] |
2407.07326
|
Andrew M Thomas
|
Andrew M. Thomas
|
Convergence of persistence diagrams for discrete time stationary processes
|
32 pages, 1 figure. Versions 1 and 2 uploaded to arXiv should be disregarded as they contained an erroneous proof of the CLT, which has been corrected in Version 3
|
Journal of Applied and Computational Topology, 9(14) (2025)
|
10.1007/s41468-025-00211-1
| null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this article we establish two fundamental results for the sublevel set persistent homology for stationary processes indexed by the positive integers. The first is a strong law of large numbers for the persistence diagram (treated as a measure "above the diagonal" in the extended plane) evaluated on a large class of sets and functions, beyond continuous functions with compact support. We prove this result subject to only minor conditions that the sequence is ergodic and the tails of the marginals are not too heavy. The second result is a central limit theorem for the persistence diagram evaluated on the class of all step functions; this result holds as long as a $\rho$-mixing criterion is satisfied and the distributions of the partial maxima do not decay too slowly. Our results greatly expand those extant in the literature to allow for more fruitful use in statistical applications, beyond idealized settings. Examples of distributions and functions for which the limit theory holds are provided throughout.
|
[
{
"version": "v1",
"created": "Wed, 10 Jul 2024 02:44:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Jul 2024 19:02:15 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Sep 2024 16:38:24 GMT"
},
{
"version": "v4",
"created": "Thu, 20 Feb 2025 21:26:24 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Thomas",
"Andrew M.",
""
]
] |
2407.07435
|
Doston Jumaniyozov
|
Doston Jumaniyozov, Surayyo Sheraliyeva
|
Infinitesmal deformations and central extensions of the $n$-th Schr\"odinger algebra
| null | null | null | null |
math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we study the second cohomology space for the $n$-th Schr\"{o}dinger algebra $\mathfrak{sch}_n$. We prove that the second cohomology space is vanishing for $n \geq 3$ and show that $\dim(H^2(\mathfrak{sch}_2, \mathfrak{sch}_2)) =2.$ Moreover, we investigate the factor algebra of the Schr\"{o}dinger algebra by its center and determine the second cohomology group of this factor algebra.
|
[
{
"version": "v1",
"created": "Wed, 10 Jul 2024 07:41:24 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 17:59:22 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jumaniyozov",
"Doston",
""
],
[
"Sheraliyeva",
"Surayyo",
""
]
] |
2407.14407
|
Vinay Kumar
|
Vinay Kumar, Claudio Cicconetti, Marco Conti, Andrea Passarella
|
Routing in Quantum Networks with End-to-End Knowledge
|
17 pages, 19 figures, Author accepted manuscript
| null | null | null |
quant-ph cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Given the diverse array of physical systems available for quantum computing and the absence of a well-defined quantum internet protocol stack, the design and optimisation of quantum networking protocols remain largely unexplored. To address this, we introduce an approach that facilitates the establishment of paths capable of delivering end-to-end fidelity above a specified threshold, without requiring detailed knowledge of the quantum network properties. In this study, we define algorithms that are specific instances of this approach and evaluate them in comparison to Dijkstra shortest path algorithm and a fully knowledge-aware algorithm through simulations. Our results demonstrate that one of the proposed algorithms consistently outperforms the other methods in delivering paths above the fidelity threshold, across various network topologies and the number of source-destination pairs involved, while maintaining significant levels of fairness among the users and being robust to inaccurate estimations of the expected end-to-end fidelity.
|
[
{
"version": "v1",
"created": "Fri, 19 Jul 2024 15:34:51 GMT"
},
{
"version": "v2",
"created": "Mon, 27 Jan 2025 17:40:12 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 08:56:42 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Kumar",
"Vinay",
""
],
[
"Cicconetti",
"Claudio",
""
],
[
"Conti",
"Marco",
""
],
[
"Passarella",
"Andrea",
""
]
] |
2407.15172
|
The CMS Collaboration
|
CMS Collaboration
|
Search for flavor-changing neutral current interactions of the top quark mediated by a Higgs boson in proton-proton collisions at 13 TeV
|
Replaced with the published version. Added the journal reference and the DOI. All the figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/TOP-22-002 (CMS Public Pages)
|
Phys. Rev. D 112 (2025) 032008
|
10.1103/95q6-vvlp
|
CMS-TOP-22-002, CERN-EP-2024-174
|
hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
A search for flavor-changing neutral current interactions of the top quark (t) and the Higgs boson (H) is presented. The search is based on proton-proton collision data collected in 2016-2018 at a center-of-mass energy of 13 TeV with the CMS detector at the LHC, and corresponding to an integrated luminosity of 138 fb$^{-1}$. Events containing a pair of leptons with the same-sign electric charge and at least one jet are considered. The results are used to constrain the branching fraction ($\mathcal{B}$) of the top quark decaying to a Higgs boson and an up (u) or charm (c) quark. No significant excess above the estimated background was found. The observed (expected) upper limits at 95% confidence level are found to be 0.072% (0.059%) for $\mathcal{B}$(t $\to$ Hu) and 0.043% (0.062%) for $\mathcal{B}$(t $\to$ Hc). These results are combined with two other searches performed by the CMS Collaboration for flavor-changing neutral current interactions of top quarks and Higgs bosons in final states where the Higgs boson decays to either a pair of photons or a pair of bottom quarks. The resulting observed (expected) upper limits at 95% confidence level are 0.019% (0.027%) for $\mathcal{B}$(t $\to$ Hu) and 0.037% (0.035%) for $\mathcal{B}$(t $\to$ Hc).
|
[
{
"version": "v1",
"created": "Sun, 21 Jul 2024 14:06:28 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 17:23:51 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"CMS Collaboration",
"",
""
]
] |
2407.16800
|
Julien Pallage
|
Julien Pallage and Antoine Lesage-Landry
|
Wasserstein Distributionally Robust Shallow Convex Neural Networks
| null | null | null | null |
cs.LG math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we propose Wasserstein distributionally robust shallow convex neural networks (WaDiRo-SCNNs) to provide reliable nonlinear predictions when subject to adverse and corrupted datasets. Our approach is based on the reformulation of a new convex training program for ReLU-based shallow neural networks, which allows us to cast the problem into the order-1 Wasserstein distributionally robust optimization framework. Our training procedure is conservative, has low stochasticity, is solvable with open-source solvers, and is scalable to large industrial deployments. We provide out-of-sample performance guarantees, show that hard convex physical constraints can be enforced in the training program, and propose a mixed-integer convex post-training verification program to evaluate model stability. WaDiRo-SCNN aims to make neural networks safer for critical applications, such as in the energy sector. Finally, we numerically demonstrate our model's performance through both a synthetic experiment and a real-world power system application, viz., the prediction of hourly energy consumption in non-residential buildings within the context of virtual power plants, and evaluate its stability across standard regression benchmark datasets. The experimental results are convincing and showcase the strengths of the proposed model.
|
[
{
"version": "v1",
"created": "Tue, 23 Jul 2024 19:01:53 GMT"
},
{
"version": "v2",
"created": "Sat, 18 Jan 2025 05:56:44 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 09:42:56 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Pallage",
"Julien",
""
],
[
"Lesage-Landry",
"Antoine",
""
]
] |
2407.20572
|
Tomohiro Tanogami
|
Tomohiro Tanogami
|
Scale locality of information flow in shell models of turbulence
|
25 pages, 5 figures. Minor revisions are made in ver. 4
| null | null | null |
cond-mat.stat-mech physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Turbulent fluctuations exhibit universal scaling laws that are independent of large-scale statistics. It is often explained that such universality is caused by the loss of information about large-scale statistics during the cascade process. In our previous study [T. Tanogami and R. Araki, Phys. Rev. Research 6, 013090 (2024)], we applied information thermodynamics to turbulence and proved that information of large-scale turbulent fluctuations is propagated to small scales. As a first step toward understanding how universality emerges at small scales under the influence of the information flow from large scales, here we investigate the scale locality of the information flow for shell models. First, we analytically show that the information flow can be decomposed into scale-local and scale-nonlocal parts. Then, by assuming the Kolmogorov hypothesis for the Kolmogorov multiplier, we prove that the scale-nonlocal part can be ignored compared to the scale-local part. This result implies that the information transfer from large to small scales occurs mainly through scale-local interactions, which is consistent with the scale locality of the energy cascade.
|
[
{
"version": "v1",
"created": "Tue, 30 Jul 2024 06:02:42 GMT"
},
{
"version": "v2",
"created": "Fri, 08 Nov 2024 03:54:25 GMT"
},
{
"version": "v3",
"created": "Tue, 08 Jul 2025 06:44:18 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 05:19:18 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Tanogami",
"Tomohiro",
""
]
] |
2407.20836
|
Naixin Zhai
|
Yunfeng Diao, Naixin Zhai, Changtao Miao, Zitong Yu, Xingxing Wei, Xun Yang, Meng Wang
|
Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks
| null | null | null | null |
cs.CV cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent advancements in image synthesis, particularly with the advent of GAN and Diffusion models, have amplified public concerns regarding the dissemination of disinformation. To address such concerns, numerous AI-generated Image (AIGI) Detectors have been proposed and achieved promising performance in identifying fake images. However, there still lacks a systematic understanding of the adversarial robustness of AIGI detectors. In this paper, we examine the vulnerability of state-of-the-art AIGI detectors against adversarial attack under white-box and black-box settings, which has been rarely investigated so far. To this end, we propose a new method to attack AIGI detectors. First, inspired by the obvious difference between real images and fake images in the frequency domain, we add perturbations under the frequency domain to push the image away from its original frequency distribution. Second, we explore the full posterior distribution of the surrogate model to further narrow this gap between heterogeneous AIGI detectors, e.g. transferring adversarial examples across CNNs and ViTs. This is achieved by introducing a novel post-train Bayesian strategy that turns a single surrogate into a Bayesian one, capable of simulating diverse victim models using one pre-trained surrogate, without the need for re-training. We name our method as Frequency-based Post-train Bayesian Attack, or FPBA. Through FPBA, we show that adversarial attack is truly a real threat to AIGI detectors, because FPBA can deliver successful black-box attacks across models, generators, defense methods, and even evade cross-generator detection, which is a crucial real-world detection scenario. The code will be shared upon acceptance.
|
[
{
"version": "v1",
"created": "Tue, 30 Jul 2024 14:07:17 GMT"
},
{
"version": "v2",
"created": "Fri, 07 Mar 2025 14:28:34 GMT"
},
{
"version": "v3",
"created": "Mon, 10 Mar 2025 06:54:28 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 14:31:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Diao",
"Yunfeng",
""
],
[
"Zhai",
"Naixin",
""
],
[
"Miao",
"Changtao",
""
],
[
"Yu",
"Zitong",
""
],
[
"Wei",
"Xingxing",
""
],
[
"Yang",
"Xun",
""
],
[
"Wang",
"Meng",
""
]
] |
2408.01390
|
Laura Pierson
|
Laura Pierson
|
Proof of a $K$-theoretic polynomial conjecture of Monical, Pechenik, and Searles
|
12 pages, comments welcome! v2: small edits to match journal version
|
Advances in Applied Mathematics, Volume 171, December 2025
|
10.1016/j.aam.2025.102959
| null |
math.CO math.KT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As part of a program to develop $K$-theoretic analogues of combinatorially important polynomials, Monical, Pechenik, and Searles (2021) proved two expansion formulas $\overline{\mathfrak{A}}_a = \sum_b Q_b^a(\beta)\overline{\mathfrak{P}}_b$ and $\overline{\mathfrak{Q}}_a = \sum_b M_b^a(\beta)\overline{\mathfrak{F}}_b,$ where each of $\overline{\mathfrak{A}}_a$, $\overline{\mathfrak{P}}_a$, $\overline{\mathfrak{Q}}_a$ and $\overline{\mathfrak{F}}_a$ is a family of polynomials that forms a basis for $\mathbb{Z}[x_1,\dots,x_n][\beta]$ indexed by weak compositions $a,$ and $Q_b^a(\beta)$ and $M_b^a(\beta)$ are monomials in $\beta$ for each pair $(a,b)$ of weak compositions. The polynomials $\overline{\mathfrak{A}}_a$ are the Lascoux atoms, $\overline{\mathfrak{P}}_a$ are the kaons, $\overline{\mathfrak{Q}}_a$ are the quasiLascoux polynomials, and $\overline{\mathfrak{F}}_a$ are the glide polynomials; these are respectively the $K$-analogues of the Demazure atoms $\mathfrak{A}_a$, the fundamental particles $\mathfrak{P}_a$, the quasikey polynomials $\mathfrak{Q}_a$, and the fundamental slide polynomials $\mathfrak{F}_a$. Monical, Pechenik, and Searles conjectured that for any fixed $a,$ $\sum_b Q_b^a(-1), \sum_b M_b^a(-1) \in \{0,1\},$ where $b$ ranges over all weak compositions. We prove this conjecture using a sign-reversing involution.
|
[
{
"version": "v1",
"created": "Fri, 02 Aug 2024 16:59:07 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 19:42:34 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 19:00:07 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Pierson",
"Laura",
""
]
] |
2408.01599
|
Minhao He
|
Minhao He, Xiaoyu Wang, Jiaqi Cai, Jonah Herzog-Arbeitman, Takashi Taniguchi, Kenji Watanabe, Ady Stern, B. Andrei Bernevig, Matthew Yankowitz, Oskar Vafek, Xiaodong Xu
|
Strongly interacting Hofstadter states in magic-angle twisted bilayer graphene
| null | null |
10.1038/s41567-025-02997-4
| null |
cond-mat.mes-hall cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Magic-angle twisted bilayer graphene (MATBG) hosts a multitude of strongly correlated states at partial fillings of its flat bands. In a magnetic field, these flat bands further evolve into a unique Hofstadter spectrum renormalized by strong Coulomb interactions. Here, we study the interacting Hofstadter states spontaneously formed within the topological magnetic subbands of an ultraclean MATBG device, notably including symmetry-broken Chern insulator (SBCI) states and fractional quantum Hall (FQH) states. The observed SBCI states form a cascade with their Chern numbers mimicking the main sequence correlated Chern insulators. The FQH states in MATBG form in Jain sequence; however, they disappear at high magnetic field, distinct from conventional FQH states which strengthen with increasing magnetic field. We reveal a unique magnetic field-driven phase transition from composite fermion phases to a dissipative Fermi liquid. Our theoretical analysis of the magnetic subbands hosting FQH states predicts non uniform quantum geometric properties far from the lowest Landau level. This points towards a more natural interpretation of these FQH states as in-field fractional Chern insulators of the magnetic subbands.
|
[
{
"version": "v1",
"created": "Fri, 02 Aug 2024 23:08:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"He",
"Minhao",
""
],
[
"Wang",
"Xiaoyu",
""
],
[
"Cai",
"Jiaqi",
""
],
[
"Herzog-Arbeitman",
"Jonah",
""
],
[
"Taniguchi",
"Takashi",
""
],
[
"Watanabe",
"Kenji",
""
],
[
"Stern",
"Ady",
""
],
[
"Bernevig",
"B. Andrei",
""
],
[
"Yankowitz",
"Matthew",
""
],
[
"Vafek",
"Oskar",
""
],
[
"Xu",
"Xiaodong",
""
]
] |
2408.03373
|
Lotta Jokiniemi
|
Daniel Castillo, Lotta Jokiniemi, Pablo Soriano, Javier Men\'endez
|
Neutrinoless $\beta\beta$ decay nuclear matrix elements complete up to N$^2$LO in heavy nuclei
|
10 pages, 4 figures, and 2 tables. Corrected a typo in Eq. (6)
|
Phys. Lett. B 860, 139181 (2025)
|
10.1016/j.physletb.2024.139181
|
INT-PUB-24-037
|
nucl-th hep-ex hep-ph nucl-ex
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We evaluate all nuclear matrix elements (NMEs) up to next-to-next-to leading order (N$^2$LO) in chiral effective field theory ($\chi$EFT) for the neutrinoless double-beta ($0\nu\beta\beta$) decay of the nuclei most relevant for experiments, including $^{76}$Ge, $^{100}$Mo, and $^{136}$Xe. We use the proton-neutron quasiparticle random-phase approximation (pnQRPA) and the nuclear shell model to calculate the N$^2$LO NMEs from very low-momentum (ultrasoft) neutrinos and from loop diagrams usually neglected in $0\nu\beta\beta$ studies. Our results indicate that the overall N$^2$LO contribution is centered around $-(5$-$10)\%$ for the shell model and $+(10$-$15)\%$ for the pnQRPA, with sizable uncertainties due to the scale dependence of the ultrasoft NMEs and the short-range nature of the loop NMEs. The sign discrepancy between many-body methods is common to all studied nuclei and points to the different behaviour of the intermediate states of the $0\nu\beta\beta$ decay. Within uncertainties, our results for the ultrasoft NME are of similar size as contributions usually referred to as ``beyond the closure approximation''.
|
[
{
"version": "v1",
"created": "Tue, 06 Aug 2024 18:00:06 GMT"
},
{
"version": "v2",
"created": "Fri, 08 Nov 2024 22:56:59 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 08:42:29 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Castillo",
"Daniel",
""
],
[
"Jokiniemi",
"Lotta",
""
],
[
"Soriano",
"Pablo",
""
],
[
"Menéndez",
"Javier",
""
]
] |
2408.06054
|
Du Nguyen
|
Du Nguyen and Stefan Sommer
|
Parallel transport on matrix manifolds and Exponential Action
| null | null | null | null |
math.NA cs.CV cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We express parallel transport for several common matrix Lie groups with a family of pseudo-Riemannian metrics in terms of matrix exponential and exponential actions. The metrics are constructed from a deformation of a bi-invariant metric and are naturally reductive. There is a similar picture for homogeneous spaces when taking quotients satisfying a general condition. In particular, for a Stiefel manifold of orthogonal matrices of size $n\times d$, we give an expression for parallel transport along a geodesic from time zero to $t$, that could be computed with time complexity of $O(n d^2)$ for small $t$, and of $O(td^3)$ for large $t$, contributing a step in a long-standing open problem in matrix manifolds. A similar result holds for {\it flag manifolds} with the canonical metric. We also show the parallel transport formulas for the {\it general linear group} and the {\it special orthogonal group} under these metrics.
|
[
{
"version": "v1",
"created": "Mon, 12 Aug 2024 11:00:04 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:10:10 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Nguyen",
"Du",
""
],
[
"Sommer",
"Stefan",
""
]
] |
2408.07686
|
Kelsey Lund
|
Kelsey A. Lund, Rahul Somasundaram, Gail C. McLaughlin, Jonah M. Miller, Matthew R. Mumpower, Ingo Tews
|
Kilonova Emissions from Neutron Star Merger Remnants: Implications for Nuclear Equation of State
|
12 pages, 4 figures
| null | null |
LA-UR-24-24836
|
astro-ph.HE nucl-th
|
http://creativecommons.org/licenses/by/4.0/
|
Multi-messenger observation of binary neutron-star mergers can provide valuable information on the nuclear equation of state (EoS). Here, we investigate to which extent electromagnetic observations of the associated kilonovae allow us to place constraints on the EoS. For this, we use state-of-the-art three-dimensional general-relativistic magneto-hydrodynamics simulations and detailed nucleosynthesis modeling to connect properties of observed light curves to properties of the accretion disk, and hence, the EoS. Using our general approach, we use multi-messenger observations of GW170817/AT2017gfo to study the impact of various sources of uncertainty on inferences of the EoS. We constrain the radius of a $\rm{1.4 M_\odot}$ neutron star to lie within $\rm{10.19\leq R_{1.4}\leq 13.0}$~km and the maximum mass to be $\rm{M_{TOV}\leq 3.06 M_\odot}$.
|
[
{
"version": "v1",
"created": "Wed, 14 Aug 2024 17:39:54 GMT"
},
{
"version": "v2",
"created": "Fri, 16 Aug 2024 18:51:25 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 01:30:26 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Lund",
"Kelsey A.",
""
],
[
"Somasundaram",
"Rahul",
""
],
[
"McLaughlin",
"Gail C.",
""
],
[
"Miller",
"Jonah M.",
""
],
[
"Mumpower",
"Matthew R.",
""
],
[
"Tews",
"Ingo",
""
]
] |
2408.10001
|
Ming Wang
|
Ming Wang, Frank Mueller
|
Coprime Bivariate Bicycle Codes and Their Layouts on Cold Atoms
| null | null | null | null |
quant-ph cs.IT math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
Quantum computing is deemed to require error correction at scale to mitigate physical noise by reducing it to lower noise levels while operating on encoded logical qubits. Popular quantum error correction schemes include CSS code, of which surface codes provide regular mappings onto 2D planes suitable for contemporary quantum devices together with known transversal logical gates. Recently, qLDPC codes have been proposed as a means to provide denser encoding with the class of bivariate bicycle (BB) codes promising feasible design for devices.
This work contributes a novel subclass of BB codes suitable for quantum error correction. This subclass employs {\em coprimes} and the product $xy$ of the two generating variables $x$ and $y$ to construct polynomials, rather than using $x$ and $y$ separately as in vanilla BB codes. In contrast to vanilla BB codes, where parameters remain unknown prior to code discovery, the rate of the proposed code can be determined beforehand by specifying a factor polynomial as an input to the numerical search algorithm. Using this coprime-BB construction, we found a number of surprisingly short to medium-length codes that were previously unknown. We also propose a layout on cold atom arrays tailored for coprime-BB codes. The proposed layout reduces both move time for short to medium-length codes and the number of moves of atoms to perform syndrome extractions. We consider an error model with global laser noise on cold atoms, and simulations show that our proposed layout achieves significant improvements over prior work across the simulated codes.
|
[
{
"version": "v1",
"created": "Mon, 19 Aug 2024 13:55:50 GMT"
},
{
"version": "v2",
"created": "Tue, 27 Aug 2024 23:34:53 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Nov 2024 21:48:14 GMT"
},
{
"version": "v4",
"created": "Sun, 29 Dec 2024 19:15:20 GMT"
},
{
"version": "v5",
"created": "Wed, 20 Aug 2025 19:23:14 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Ming",
""
],
[
"Mueller",
"Frank",
""
]
] |
2408.10262
|
Padraig Cunningham Prof
|
Padraig Cunningham, Barry Smyth
|
An Analysis of the Impact of Gold Open Access Publications in Computer Science
|
13 pages, 8 figures
| null |
10.1145/3721975
| null |
cs.DL
|
http://creativecommons.org/licenses/by/4.0/
|
There has been some concern about the impact of predatory publishers on scientific research for some time. Recently, publishers that might previously have been considered `predatory' have established their bona fides, at least to the extent that they are included in citation impact scores such as the field-weighted citation impact (FWCI). These are sometimes called `grey' publishers (MDPI, Frontiers, Hindawi). In this paper, we show that the citation landscape for these grey publications is significantly different from the mainstream landscape and that affording publications in these venues the same status as publications in mainstream journals may significantly distort metrics such as the FWCI.
|
[
{
"version": "v1",
"created": "Thu, 15 Aug 2024 15:50:32 GMT"
},
{
"version": "v2",
"created": "Sat, 07 Dec 2024 15:16:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Cunningham",
"Padraig",
""
],
[
"Smyth",
"Barry",
""
]
] |
2408.10264
|
Gefei Shen
|
Chengyu Gong, Gefei Shen, Luanzheng Guo, Nathan Tallent, Dongfang Zhao
|
OPDR: Order-Preserving Dimension Reduction for Semantic Embedding of Multimodal Scientific Data
| null | null | null | null |
cs.LG cs.AI cs.IR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
One of the most common operations in multimodal scientific data management is searching for the $k$ most similar items (or, $k$-nearest neighbors, KNN) from the database after being provided a new item. Although recent advances of multimodal machine learning models offer a \textit{semantic} index, the so-called \textit{embedding vectors} mapped from the original multimodal data, the dimension of the resulting embedding vectors are usually on the order of hundreds or a thousand, which are impractically high for time-sensitive scientific applications.
This work proposes to reduce the dimensionality of the output embedding vectors such that the set of top-$k$ nearest neighbors do not change in the lower-dimensional space, namely Order-Preserving Dimension Reduction (OPDR). In order to develop such an OPDR method, our central hypothesis is that by analyzing the intrinsic relationship among key parameters during the dimension-reduction map, a quantitative function may be constructed to reveal the correlation between the target (lower) dimensionality and other variables. To demonstrate the hypothesis, this paper first defines a formal measure function to quantify the KNN similarity for a specific vector, then extends the measure into an aggregate accuracy of the global metric spaces, and finally derives a closed-form function between the target (lower) dimensionality and other variables. We incorporate the closed-function into popular dimension-reduction methods, various distance metrics, and embedding models.
|
[
{
"version": "v1",
"created": "Thu, 15 Aug 2024 22:30:44 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:25:23 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Gong",
"Chengyu",
""
],
[
"Shen",
"Gefei",
""
],
[
"Guo",
"Luanzheng",
""
],
[
"Tallent",
"Nathan",
""
],
[
"Zhao",
"Dongfang",
""
]
] |
2408.10889
|
Alberto Pozanco
|
Marianela Morales, Alberto Pozanco, Giuseppe Canonaco, Sriram Gopalakrishnan, Daniel Borrajo, Manuela Veloso
|
On Learning Action Costs from Input Plans
| null | null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Most of the work on learning action models focus on learning the actions' dynamics from input plans. This allows us to specify the valid plans of a planning task. However, very little work focuses on learning action costs, which in turn allows us to rank the different plans. In this paper we introduce a new problem: that of learning the costs of a set of actions such that a set of input plans are optimal under the resulting planning model. To solve this problem we present $LACFIP^k$, an algorithm to learn action's costs from unlabeled input plans. We provide theoretical and empirical results showing how $LACFIP^k$ can successfully solve this task.
|
[
{
"version": "v1",
"created": "Tue, 20 Aug 2024 14:20:19 GMT"
},
{
"version": "v2",
"created": "Mon, 02 Sep 2024 09:48:43 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 10:39:45 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Morales",
"Marianela",
""
],
[
"Pozanco",
"Alberto",
""
],
[
"Canonaco",
"Giuseppe",
""
],
[
"Gopalakrishnan",
"Sriram",
""
],
[
"Borrajo",
"Daniel",
""
],
[
"Veloso",
"Manuela",
""
]
] |
2408.11896
|
Urs Schreiber
|
Hisham Sati, Urs Schreiber
|
Cohomotopy, Framed Links, and Abelian Anyons
|
18 pages; v2: new title, abstract, intro and outro, aimed more at pure mathematicians
| null | null | null |
hep-th cond-mat.mes-hall math-ph math.AT math.MP quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We establish a natural identification of cobordism classes of framed links with the fundamental group of the group-completed configuration space of points in the plane, by appeal to Okuyama's previously underappreciated interval configuration model for the latter. Under Segal's theorem, these classes are integers generated by the Hopf generator in the 2-cohomotopy of the 3-sphere, and we identify these knot-theoretically with the writhe sum of the linking and framing number of links. We observe that this link invariant is the properly regularized Wilson line observable of abelian Chern-Simons theory, and show, in consequence of the main theorem, that this arises equivalently as the expectation value of pure quantum states on the group algebra, under link sum, of cobordism classes of framed links. Observing that these quantum states regard framed links as worldlines of anyons in that they assign a fixed complex phase factor to each crossing (braiding) of strands, we close with an outlook on implications for the identification of anyonic solitons in 2D electron gases as exotic flux quantized in Cohomotopy instead of in ordinary cohomology.
|
[
{
"version": "v1",
"created": "Wed, 21 Aug 2024 18:00:02 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 20:46:55 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Sati",
"Hisham",
""
],
[
"Schreiber",
"Urs",
""
]
] |
2408.13003
|
Vukasin Stanojevic
|
Vuka\v{s}in Stanojevi\'c and Branimir Todorovi\'c
|
BoostTrack++: using tracklet information to detect more objects in multiple object tracking
|
To be published in Filomat, Vol 39, No 16 (2025)
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Multiple object tracking (MOT) depends heavily on selection of true positive detected bounding boxes. However, this aspect of the problem is mostly overlooked or mitigated by employing two-stage association and utilizing low confidence detections in the second stage. Recently proposed BoostTrack attempts to avoid the drawbacks of multiple stage association approach and use low-confidence detections by applying detection confidence boosting. In this paper, we identify the limitations of the confidence boost used in BoostTrack and propose a method to improve its performance. To construct a richer similarity measure and enable a better selection of true positive detections, we propose to use a combination of shape, Mahalanobis distance and novel soft BIoU similarity. We propose a soft detection confidence boost technique which calculates new confidence scores based on the similarity measure and the previous confidence scores, and we introduce varying similarity threshold to account for lower similarity measure between detections and tracklets which are not regularly updated. The proposed additions are mutually independent and can be used in any MOT algorithm.
Combined with the BoostTrack+ baseline, our method achieves near state of the art results on the MOT17 dataset and new state of the art HOTA and IDF1 scores on the MOT20 dataset.
The source code is available at: https://github.com/vukasin-stanojevic/BoostTrack .
|
[
{
"version": "v1",
"created": "Fri, 23 Aug 2024 11:44:21 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 22:40:59 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Stanojević",
"Vukašin",
""
],
[
"Todorović",
"Branimir",
""
]
] |
2408.13802
|
Xiongwei Zhao
|
Xiongwei Zhao, Congcong Wen, Xu Zhu, Yang Wang, Haojie Bai and Wenhao Dou
|
TripleMixer: A 3D Point Cloud Denoising Model for Adverse Weather
|
15 pages, submit to IEEE TIP
| null | null | null |
cs.CV cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Adverse weather conditions such as snow, fog, and rain pose significant challenges to LiDAR-based perception models by introducing noise and corrupting point cloud measurements. To address this issue, we propose TripleMixer, a robust and efficient point cloud denoising network that integrates spatial, frequency, and channel-wise processing through three specialized mixer modules. TripleMixer effectively suppresses high-frequency noise while preserving essential geometric structures and can be seamlessly deployed as a plug-and-play module within existing LiDAR perception pipelines. To support the development and evaluation of denoising methods, we construct two large-scale simulated datasets, Weather-KITTI and Weather-NuScenes, covering diverse weather scenarios with dense point-wise semantic and noise annotations. Based on these datasets, we establish four benchmarks: Denoising, Semantic Segmentation (SS), Place Recognition (PR), and Object Detection (OD). These benchmarks enable systematic evaluation of denoising generalization, transferability, and downstream impact under both simulated and real-world adverse weather conditions. Extensive experiments demonstrate that TripleMixer achieves state-of-the-art denoising performance and yields substantial improvements across all downstream tasks without requiring retraining. Our results highlight the potential of denoising as a task-agnostic preprocessing strategy to enhance LiDAR robustness in real-world autonomous driving applications.
|
[
{
"version": "v1",
"created": "Sun, 25 Aug 2024 10:45:52 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:05:53 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Zhao",
"Xiongwei",
""
],
[
"Wen",
"Congcong",
""
],
[
"Zhu",
"Xu",
""
],
[
"Wang",
"Yang",
""
],
[
"Bai",
"Haojie",
""
],
[
"Dou",
"Wenhao",
""
]
] |
2408.14602
|
Xincheng Lin
|
Xincheng Lin, Jared Vanasse
|
Two-Body Triton Photodisintegration and Wigner-SU(4) Symmetry
|
40 pages, 6 figure; Published version
|
Phys. Rev. C 112, 024001, 2025
|
10.1103/qgcx-r63r
| null |
nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We calculate the two-body triton photodisintegration cross section as a function of photon energy to next-to-next-to leading order (NNLO) in pionless effective field theory (EFT($\pi \!\!/$)) and show good agreement with experiment. In addition we calculate the polarization asymmetry $R_C=-0.441(15)$ in cold neutron-deuteron capture to NNLO in EFT($\pi \!\!/$), in agreement with the experimental value of $R_C=-0.42\pm 0.03$ [M. W. Konijnenberg et al. in Phys. Lett. B 205, 215 (1988)]. We also assess the dependence of $R_C$ on different fits of the two-nucleon magnetic currents. Finally, we consider the impact of Wigner-SU(4) symmetry and demonstrate that starting from the Wigner-SU(4) symmetric limit and including perturbative corrections to the breaking of Wigner-SU(4) symmetry does a good job of describing two-body triton photodisintegration.
|
[
{
"version": "v1",
"created": "Mon, 26 Aug 2024 19:47:09 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 02:09:18 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Lin",
"Xincheng",
""
],
[
"Vanasse",
"Jared",
""
]
] |
2408.17401
|
Antonio Rago
|
Antonio Rago, Bence Palfi, Purin Sukpanichnant, Hannibal Nabli, Kavyesh Vivek, Olga Kostopoulou, James Kinross and Francesca Toni
|
Exploring the Effect of Explanation Content and Format on User Comprehension and Trust in Healthcare
|
12 pages. To be published at The 14th Conference on Prestigious Applications of Intelligent Systems (PAIS-2025)
| null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
AI-driven tools for healthcare are widely acknowledged as potentially beneficial to health practitioners and patients, e.g. the QCancer regression tool for cancer risk prediction. However, for these tools to be trusted, they need to be supplemented with explanations. We examine how explanations' content and format affect user comprehension and trust when explaining QCancer's predictions. Regarding content, we deploy the SHAP and Occlusion-1 explanation methods. Regarding format, we present SHAP explanations, conventionally, as charts (SC) and Occlusion-1 explanations as charts (OC) as well as text (OT), to which their simpler nature lends itself. We conduct experiments with two sets of stakeholders: the general public (representing patients) and medical students (representing healthcare practitioners). Our experiments showed higher subjective comprehension and trust for Occlusion-1 over SHAP explanations based on content. However, when controlling for format, only OT outperformed SC, suggesting this trend is driven by preferences for text. Other findings corroborated that explanation format, rather than content, is often the critical factor.
|
[
{
"version": "v1",
"created": "Fri, 30 Aug 2024 16:36:53 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Feb 2025 17:49:36 GMT"
},
{
"version": "v3",
"created": "Tue, 20 May 2025 16:29:43 GMT"
},
{
"version": "v4",
"created": "Thu, 21 Aug 2025 15:00:51 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Rago",
"Antonio",
""
],
[
"Palfi",
"Bence",
""
],
[
"Sukpanichnant",
"Purin",
""
],
[
"Nabli",
"Hannibal",
""
],
[
"Vivek",
"Kavyesh",
""
],
[
"Kostopoulou",
"Olga",
""
],
[
"Kinross",
"James",
""
],
[
"Toni",
"Francesca",
""
]
] |
2409.02998
|
Giulia Fardelli
|
Giulia Fardelli, A. Liam Fitzpatrick, Emanuel Katz
|
Constructing the Infrared Conformal Generators on the Fuzzy Sphere
|
31 pages, 13 figures; v2: minor corrections, journal version
| null |
10.21468/SciPostPhys.18.3.086
| null |
hep-th cond-mat.stat-mech
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the conformal algebra on the fuzzy sphere, and in particular the generators of translations and special conformal transformations which are emergent symmetries in the infinite IR but are broken along the RG flow. We show how to extract these generators using the energy momentum tensor, which is complicated by the fact that one does not have a priori access to the energy momentum tensor of the CFT limit but rather must construct it numerically. We discuss and quantitatively analyze the main sources of corrections to the conformal generators due to the breaking of scale-invariance at finite energy, and develop efficient methods for removing these corrections. The resulting generators have matrix elements that match CFT predictions with accuracy varying from sub-percent level for the lowest-lying states up to several percent accuracy for states with dimension $\sim 5$ with $N=16$ fermions. We show that the generators can be used to accurately identify primary operators vs descendant operators in energy ranges where the spectrum is too dense to do the identification solely based on the approximate integer spacing within conformal multiplets.
|
[
{
"version": "v1",
"created": "Wed, 04 Sep 2024 18:00:09 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 13:50:59 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fardelli",
"Giulia",
""
],
[
"Fitzpatrick",
"A. Liam",
""
],
[
"Katz",
"Emanuel",
""
]
] |
2409.04215
|
Anna Broms Ph.D.
|
Anna Broms, Alex H. Barnett and Anna-Karin Tornberg
|
A Method of Fundamental Solutions for Large-Scale 3D Elastance and Mobility Problems
|
27 pages, 11 figures, accepted for publication in Advances in Computational Mathematics
| null | null | null |
math.NA cs.NA physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
The method of fundamental solutions (MFS) is known to be effective for solving 3D Laplace and Stokes Dirichlet boundary value problems in the exterior of a large collection of simple smooth objects. Here we present new scalable MFS formulations for the corresponding elastance and mobility problems. The elastance problem computes the potentials of conductors with given net charges, while the mobility problem -- crucial to rheology and complex fluid applications -- computes rigid body velocities given net forces and torques on the particles. The key idea is orthogonal projection of the net charge (or forces and torques) in a rectangular variant of a "completion flow". The proposal is compatible with one-body preconditioning, resulting in well-conditioned square linear systems amenable to fast multipole accelerated iterative solution, thus a cost linear in the particle number. For large suspensions with moderate lubrication forces, MFS sources on inner proxy-surfaces give accuracy on par with a well-resolved boundary integral formulation. Our several numerical tests include a suspension of 10000 nearby ellipsoids, using 26 million total preconditioned degrees of freedom, where GMRES converges to five digits of accuracy in under two hours on one workstation.
|
[
{
"version": "v1",
"created": "Fri, 06 Sep 2024 12:12:42 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 15:29:08 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Broms",
"Anna",
""
],
[
"Barnett",
"Alex H.",
""
],
[
"Tornberg",
"Anna-Karin",
""
]
] |
2409.05500
|
Ce Guo
|
Ziyang Jiao, Ce Guo and Wayne Luk
|
Scalable Time-Series Causal Discovery with Approximate Causal Ordering
| null | null | null | null |
cs.LG cs.DC cs.PF stat.CO
|
http://creativecommons.org/licenses/by/4.0/
|
Causal discovery in time-series data presents a significant computational challenge. Standard algorithms are often prohibitively expensive for datasets with many variables or samples. This study introduces and validates a heuristic approximation of the VarLiNGAM algorithm to address this scalability problem. The standard VarLiNGAM method relies on an iterative search, recalculating statistical dependencies after each step. Our heuristic modifies this procedure by omitting the iterative refinement. This change permits a one-time precomputation of all necessary statistical values. The algorithmic modification reduces the time complexity from $O(m^3n)$ to $O(m^2n + m^3)$ while keeping the space complexity at $O(m^2)$, where $m$ is the number of variables and $n$ is the number of samples. While an approximation, our approach retains VarLiNGAM's essential structure and empirical reliability. On large-scale financial data with up to 400 variables, our algorithm achieves a 7--13x speedup over the standard implementation and a 4.5x speedup over a GPU-accelerated version. Evaluations across medical imaging, web server monitoring, and finance demonstrate the heuristic's robustness and practical scalability. This work offers a validated balance between computational efficiency and discovery quality, making large-scale causal analysis feasible on personal computers.
|
[
{
"version": "v1",
"created": "Mon, 09 Sep 2024 10:52:58 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Sep 2024 08:01:24 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 14:56:53 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Jiao",
"Ziyang",
""
],
[
"Guo",
"Ce",
""
],
[
"Luk",
"Wayne",
""
]
] |
2409.06482
|
Fernando Parisio
|
Fernando Parisio
|
Quantum-state texture and gate identification
|
To appear on Phys. Rev. Lett
| null |
10.1103/PhysRevLett.133.260801
| null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce and explore the notion of texture of an arbitrary quantum state, in a selected basis. In the first part of this letter we develop a resource theory and show that state texture is adequately described by an easily computable monotone, which is also directly measurable. It is shown that textures are useful in the characterization of unknown quantum gates in universal circuit layers. By using randomized input states and recording the textures of the output qubits we are able to fully characterize the circuit layer, whenever it contains at least one CNOT gate. This can be done without the need of tomographic protocols and the use of ancillary systems.
|
[
{
"version": "v1",
"created": "Tue, 10 Sep 2024 13:11:34 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Nov 2024 19:39:26 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Parisio",
"Fernando",
""
]
] |
2409.06922
|
Jonathan Stanfill
|
Guglielmo Fucci, Mateusz Piorkowski, Jonathan Stanfill
|
The spectral $\zeta$-function for quasi-regular Sturm--Liouville operators
|
39 pages
|
Lett Math Phys 115, 8 (2025)
|
10.1007/s11005-024-01893-x
| null |
math-ph math.MP math.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this work we analyze the spectral $\zeta$-function associated with the self-adjoint extensions, $T_{A,B}$, of quasi-regular Sturm--Liouville operators that are bounded from below. By utilizing the Green's function formalism, we find the characteristic function which implicitly provides the eigenvalues associated with a given self-adjoint extension $T_{A,B}$. The characteristic function is then employed to construct a contour integral representation for the spectral $\zeta$-function of $T_{A,B}$. By assuming a general form for the asymptotic expansion of the characteristic function, we describe the analytic continuation of the $\zeta$-function to a larger region of the complex plane. We also present a method for computing the value of the spectral $\zeta$-function of $T_{A,B}$ at all positive integers. We provide two examples to illustrate the methods developed in the paper: the generalized Bessel and Legendre operators. We show that in the case of the generalized Bessel operator, the spectral $\zeta$-function develops a branch point at the origin, while in the case of the Legendre operator it presents, more remarkably, branch points at every nonpositive integer value of $s$.
|
[
{
"version": "v1",
"created": "Wed, 11 Sep 2024 00:38:11 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fucci",
"Guglielmo",
""
],
[
"Piorkowski",
"Mateusz",
""
],
[
"Stanfill",
"Jonathan",
""
]
] |
2409.12892
|
Lukas H\"ollein
|
Lukas H\"ollein, Alja\v{z} Bo\v{z}i\v{c}, Michael Zollh\"ofer, Matthias Nie{\ss}ner
|
3DGS-LM: Faster Gaussian-Splatting Optimization with Levenberg-Marquardt
|
Accepted to ICCV 2025. Project page: https://lukashoel.github.io/3DGS-LM, Video: https://www.youtube.com/watch?v=tDiGuGMssg8, Code: https://github.com/lukasHoel/3DGS-LM
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present 3DGS-LM, a new method that accelerates the reconstruction of 3D Gaussian Splatting (3DGS) by replacing its ADAM optimizer with a tailored Levenberg-Marquardt (LM). Existing methods reduce the optimization time by decreasing the number of Gaussians or by improving the implementation of the differentiable rasterizer. However, they still rely on the ADAM optimizer to fit Gaussian parameters of a scene in thousands of iterations, which can take up to an hour. To this end, we change the optimizer to LM that runs in conjunction with the 3DGS differentiable rasterizer. For efficient GPU parallization, we propose a caching data structure for intermediate gradients that allows us to efficiently calculate Jacobian-vector products in custom CUDA kernels. In every LM iteration, we calculate update directions from multiple image subsets using these kernels and combine them in a weighted mean. Overall, our method is 20% faster than the original 3DGS while obtaining the same reconstruction quality. Our optimization is also agnostic to other methods that acclerate 3DGS, thus enabling even faster speedups compared to vanilla 3DGS.
|
[
{
"version": "v1",
"created": "Thu, 19 Sep 2024 16:31:44 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 12:52:11 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Höllein",
"Lukas",
""
],
[
"Božič",
"Aljaž",
""
],
[
"Zollhöfer",
"Michael",
""
],
[
"Nießner",
"Matthias",
""
]
] |
2409.13644
|
Qifeng Hu
|
Qifeng Hu, Shamsulhaq Basir, Inanc Senocak
|
Non-overlapping, Schwarz-type Domain Decomposition Method for Physics and Equality Constrained Artificial Neural Networks
|
49 pages, 19 figures
| null | null | null |
cs.LG math.OC physics.comp-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present a non-overlapping, Schwarz-type domain decomposition method with a generalized interface condition, designed for physics-informed machine learning of partial differential equations (PDEs) in both forward and inverse contexts. Our approach employs physics and equality-constrained artificial neural networks (PECANN) within each subdomain. Unlike the original PECANN method, which relies solely on initial and boundary conditions to constrain PDEs, our method uses both boundary conditions and the governing PDE to constrain a unique interface loss function for each subdomain. This modification improves the learning of subdomain-specific interface parameters while reducing communication overhead by delaying information exchange between neighboring subdomains. To address the constrained optimization in each subdomain, we apply an augmented Lagrangian method with a conditionally adaptive update strategy, transforming the problem into an unconstrained dual optimization. A distinct advantage of our domain decomposition method is its ability to learn solutions to both Poisson's and Helmholtz equations, even in cases with high-wavenumber and complex-valued solutions. Through numerical experiments with up to 64 subdomains, we demonstrate that our method consistently generalizes well as the number of subdomains increases.
|
[
{
"version": "v1",
"created": "Fri, 20 Sep 2024 16:48:55 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Nov 2024 01:23:55 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hu",
"Qifeng",
""
],
[
"Basir",
"Shamsulhaq",
""
],
[
"Senocak",
"Inanc",
""
]
] |
2409.13792
|
Nilay Kushawaha
|
Nilay Kushawaha, Egidio Falotico
|
Continual Learning for Multimodal Data Fusion of a Soft Gripper
|
Accepted in Wiley Advanced Robotics Research
| null | null | null |
cs.LG cs.AI cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Continual learning (CL) refers to the ability of an algorithm to continuously and incrementally acquire new knowledge from its environment while retaining previously learned information. A model trained on one data modality often fails when tested with a different modality. A straightforward approach might be to fuse the two modalities by concatenating their features and training the model on the fused data. However, this requires retraining the model from scratch each time it encounters a new domain. In this paper, we introduce a continual learning algorithm capable of incrementally learning different data modalities by leveraging both class-incremental and domain-incremental learning scenarios in an artificial environment where labeled data is scarce, yet non-iid (independent and identical distribution) unlabeled data from the environment is plentiful. The proposed algorithm is efficient and only requires storing prototypes for each class. We evaluate the algorithm's effectiveness on a challenging custom multimodal dataset comprising of tactile data from a soft pneumatic gripper, and visual data from non-stationary images of objects extracted from video sequences. Additionally, we conduct an ablation study on the custom dataset and the Core50 dataset to highlight the contributions of different components of the algorithm. To further demonstrate the robustness of the algorithm, we perform a real-time experiment for object classification using the soft gripper and an external independent camera setup, all synchronized with the Robot Operating System (ROS) framework.
|
[
{
"version": "v1",
"created": "Fri, 20 Sep 2024 09:53:27 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 07:09:11 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Kushawaha",
"Nilay",
""
],
[
"Falotico",
"Egidio",
""
]
] |
2409.14089
|
Valeria Repetto
|
Valeria Repetto, Elia Giuseppe Ceroni, Giuseppe Buonaiuto, Romina D'Aurizio
|
Quantum enhanced stratification of Breast Cancer: exploring quantum expressivity for real omics data
|
10 pages, 6 figures
|
Quantum Machine Intelligence 2025
|
10.1007/s42484-025-00289-x
|
81
|
quant-ph cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum Machine Learning (QML) is considered one of the most promising applications of Quantum Computing in the Noisy Intermediate Scale Quantum (NISQ) era for the impact it is thought to have in the near future. Although promising theoretical assumptions, the exploration of how QML could foster new discoveries in Medicine and Biology fields is still in its infancy with few examples. In this study, we aimed to assess whether Quantum Kernels (QK) could effectively classify subtypes of Breast Cancer (BC) patients on the basis of molecular characteristics. We performed an heuristic exploration of encoding configurations with different entanglement levels to determine a trade-off between kernel expressivity and performances. Our results show that QKs yield comparable clustering results with classical methods while using fewer data points, and are able to fit the data with a higher number of clusters. Additionally, we conducted the experiments on the Quantum Processing Unit (QPU) to evaluate the effect of noise on the outcome. We found that less expressive encodings showed a higher resilience to noise, indicating that the computational pipeline can be reliably implemented on the NISQ devices. Our findings suggest that QK methods show promises for application in Precision Oncology, especially in scenarios where the dataset is limited in size and a granular non-trivial stratification of complex molecular data cannot be achieved classically.
|
[
{
"version": "v1",
"created": "Sat, 21 Sep 2024 10:00:09 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Jul 2025 11:47:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Repetto",
"Valeria",
""
],
[
"Ceroni",
"Elia Giuseppe",
""
],
[
"Buonaiuto",
"Giuseppe",
""
],
[
"D'Aurizio",
"Romina",
""
]
] |
2409.14487
|
Boqu Chen
|
Boqu Chen, Ce Li, Xiaoxuan Li, Ding Zhao, Lu Cai, Kaikai Du and Min Qiu
|
Ultra-Thin, Ultra-Light, Rainbow-Free AR Glasses Based on Single-Layer Full-Color SiC Diffrcative Waveguide
| null |
eLight 5, 21 (2025)
|
10.1186/s43593-025-00100-1
| null |
physics.optics
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As information interaction technology advances, the efficiency, dimensionality, and user experience of information transmission have significantly improved. Communication has evolved from letters to telegraphs, markedly increasing transmission speed; from telephones to video calls, enhancing communication dimensions; and from smartphones to augmented reality (AR) displays, which provide increasingly immersive user experiences. Surface relief grating (SRG) diffractive waveguides have attracted considerable attention for their optimal balance between weight, size, optical performance, and mass production capabilities, positioning them as a leading solution for AR displays. However, as consumer expectations for higher display quality and better device integration rise, traditional high-refractive-index glass-based diffractive waveguides face limitations, including bulkiness, heavy weight, and conspicuous rainbow artifacts in full-color displays. To overcome these challenges, a novel solution: ultra-thin, lightweight silicon carbide (SiC) AR prescription glasses was proposed. This solution achieves full-color displays without rainbow artifacts, with total weight of just 2.685 g and thickness of only 0.55 mm. Moreover, these glasses are compatible with prescription Fresnel lenses and are well-suited for scalable mass production. This innovation provides a robust platform for the seamless integration of augmented reality into daily life, offering significant potential to enhance user interaction.
|
[
{
"version": "v1",
"created": "Sun, 22 Sep 2024 15:17:08 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Chen",
"Boqu",
""
],
[
"Li",
"Ce",
""
],
[
"Li",
"Xiaoxuan",
""
],
[
"Zhao",
"Ding",
""
],
[
"Cai",
"Lu",
""
],
[
"Du",
"Kaikai",
""
],
[
"Qiu",
"Min",
""
]
] |
2409.14815
|
Jonathan K\"ulz
|
Daniel Ostermeier and Jonathan K\"ulz and Matthias Althoff
|
Automatic Geometric Decomposition for Analytical Inverse Kinematics
|
Website: https://eaik.cps.cit.tum.de/
|
IEEE Robotics and Automation Letters (RA-L), Vol. 10, No. 10, pp. 9964--9971, 2025
|
10.1109/LRA.2025.3597897
| null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Calculating the inverse kinematics (IK) is a fundamental challenge in robotics. Compared to numerical or learning-based approaches, analytical IK provides higher efficiency and accuracy. However, existing analytical approaches are difficult to use in most applications, as they require human ingenuity in the derivation process, are numerically unstable, or rely on time-consuming symbolic manipulation. In contrast, we propose a method that, for the first time, enables an analytical IK derivation and computation in less than a millisecond in total. Our work is based on an automatic online decomposition of the IK into pre-solved, numerically stable subproblems via a kinematic classification of the respective manipulator. In numerical experiments, we demonstrate that our approach is orders of magnitude faster in deriving the IK than existing tools that employ symbolic manipulation. Following this one-time derivation, our method matches and often surpasses baselines, such as IKFast, in terms of speed and accuracy during the computation of explicit IK solutions. Finally, we provide an open-source C++ toolbox with Python wrappers that substantially reduces the entry barrier to using analytical IK in applications like rapid prototyping and kinematic robot design.
|
[
{
"version": "v1",
"created": "Mon, 23 Sep 2024 08:45:40 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:54:47 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ostermeier",
"Daniel",
""
],
[
"Külz",
"Jonathan",
""
],
[
"Althoff",
"Matthias",
""
]
] |
2409.16531
|
Michael Plank
|
M. J. Plank, W. S. Hart, J. Polonsky, M. Keita, S. Ahuka-Mundeke, R. N. Thompson
|
Estimation of end-of-outbreak probabilities in the presence of delayed and incomplete case reporting
| null |
Proc. R. Soc. B (2025) 292: 20242825
|
10.1098/rspb.2024.2825
| null |
q-bio.PE q-bio.QM
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Towards the end of an infectious disease outbreak, when a period has elapsed without new case notifications, a key question for public health policy makers is whether the outbreak can be declared over. This requires the benefits of a declaration (e.g., relaxation of outbreak control measures) to be balanced against the risk of a resurgence in cases. To support this decision making, mathematical methods have been developed to quantify the end-of-outbreak probability. Here, we propose a new approach to this problem that accounts for a range of features of real-world outbreaks, specifically: (i) incomplete case ascertainment; (ii) reporting delays; (iii) individual heterogeneity in transmissibility; and (iv) whether cases were imported or infected locally. We showcase our approach using two case studies: Covid-19 in New Zealand in 2020, and Ebola virus disease in the Democratic Republic of the Congo in 2018. In these examples, we found that the date when the estimated probability of no future infections reached 95% was relatively consistent across a range of modelling assumptions. This suggests that our modelling framework can generate robust quantitative estimates that can be used by policy advisors, alongside other sources of evidence, to inform end-of-outbreak declarations.
|
[
{
"version": "v1",
"created": "Wed, 25 Sep 2024 00:56:34 GMT"
},
{
"version": "v2",
"created": "Sun, 24 Nov 2024 23:20:32 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Plank",
"M. J.",
""
],
[
"Hart",
"W. S.",
""
],
[
"Polonsky",
"J.",
""
],
[
"Keita",
"M.",
""
],
[
"Ahuka-Mundeke",
"S.",
""
],
[
"Thompson",
"R. N.",
""
]
] |
2409.20135
|
Zezhou Wang
|
Zezhou Wang, Yaxin Du, Xingjun Ma, Yugang Jiang, Zhuzhong Qian, Siheng Chen
|
Optimizing Cross-Client Domain Coverage for Federated Instruction Tuning of Large Language Models
|
EMNLP 2025
| null | null | null |
cs.LG cs.CL cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Federated domain-specific instruction tuning (FedDIT) for large language models (LLMs) aims to enhance performance in specialized domains using distributed private and limited data, yet identifying key performance drivers and optimal augmentation strategies remains challenging. We empirically establish that cross-client domain coverage, rather than data heterogeneity, is the pivotal factor. We then introduce FedDCA, an algorithm that explicitly maximizes this coverage through diversity-oriented client center selection and retrieval-based augmentation, constructing diverse, non-redundant cross-client instruction sets. Extensive experiments across multiple domains demonstrate FedDCA's superiority over eleven baselines, achieving performance gains of up to 29.19\% and domain coverage improvements of 4.82\%-21.36\%. FedDCA maintains its effectiveness in diverse and challenging scenarios, including data selection, held-out settings where task-specific public data is scarce and various data heterogeneity, with manageable privacy risks. This work clarifies critical FedDIT dynamics and presents FedDCA as an effective, privacy-preserving, and scalable solution for advancing domain-specific LLM tuning.
|
[
{
"version": "v1",
"created": "Mon, 30 Sep 2024 09:34:31 GMT"
},
{
"version": "v2",
"created": "Tue, 01 Oct 2024 05:37:07 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Oct 2024 08:32:02 GMT"
},
{
"version": "v4",
"created": "Fri, 11 Oct 2024 12:19:57 GMT"
},
{
"version": "v5",
"created": "Tue, 21 Jan 2025 09:25:25 GMT"
},
{
"version": "v6",
"created": "Thu, 21 Aug 2025 09:56:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Zezhou",
""
],
[
"Du",
"Yaxin",
""
],
[
"Ma",
"Xingjun",
""
],
[
"Jiang",
"Yugang",
""
],
[
"Qian",
"Zhuzhong",
""
],
[
"Chen",
"Siheng",
""
]
] |
2409.20173
|
Petr Vanc
|
Petr Vanc, Giovanni Franzese, Jan Kristof Behrens, Cosimo Della Santina, Karla Stepanova, Jens Kober, Robert Babuska
|
ILeSiA: Interactive Learning of Robot Situational Awareness from Camera Input
|
8 pages, 9 figures. Accepted to IEEE Robotics and Automation Letters (Early Access)
| null |
10.1109/LRA.2025.3601037
| null |
cs.RO cs.CV cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Learning from demonstration is a promising approach for teaching robots new skills. However, a central challenge in the execution of acquired skills is the ability to recognize faults and prevent failures. This is essential because demonstrations typically cover only a limited set of scenarios and often only the successful ones. During task execution, unforeseen situations may arise, such as changes in the robot's environment or interaction with human operators. To recognize such situations, this paper focuses on teaching the robot situational awareness by using a camera input and labeling frames as safe or risky. We train a Gaussian Process (GP) regression model fed by a low-dimensional latent space representation of the input images. The model outputs a continuous risk score ranging from zero to one, quantifying the degree of risk at each timestep. This allows for pausing task execution in unsafe situations and directly adding new training data, labeled by the human user. Our experiments on a robotic manipulator show that the proposed method can reliably detect both known and novel faults using only a single example for each new fault. In contrast, a standard multi-layer perceptron (MLP) performs well only on faults it has encountered during training. Our method enables the next generation of cobots to be rapidly deployed with easy-to-set-up, vision-based risk assessment, proactively safeguarding humans and detecting misaligned parts or missing objects before failures occur. We provide all the code and data required to reproduce our experiments at imitrob.ciirc.cvut.cz/publications/ilesia.
|
[
{
"version": "v1",
"created": "Mon, 30 Sep 2024 10:31:27 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 13:29:27 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Vanc",
"Petr",
""
],
[
"Franzese",
"Giovanni",
""
],
[
"Behrens",
"Jan Kristof",
""
],
[
"Della Santina",
"Cosimo",
""
],
[
"Stepanova",
"Karla",
""
],
[
"Kober",
"Jens",
""
],
[
"Babuska",
"Robert",
""
]
] |
2410.00383
|
Z. Yousaf
|
Z. Yousaf, Kazuharu Bamba, M. Z. Bhatti and S. Khan
|
Completely Deformed Complexity-free Anisotropic Fluid Sphere
|
18 pages, 6 figures, title changed, further investigations added
| null | null |
FU-PCG-141
|
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this work, we investigate the emergence of compact, anisotropic stellar structures through the gravitational decoupling scheme within the framework of complete geometric deformation. The study introduces a novel synthesis of two independent techniques, namely the zero-complexity factor and density-like constraints, applied simultaneously to determine the deformation functions. This dual implementation represents a new methodological step in stellar modeling, as it allows us to explicitly control the role of anisotropy and complexity in the internal structure of self-gravitating objects. Starting from a chosen metric ansatz as a seed solution, we demonstrate that the zero-complexity condition captures the gravitational response of compact matter in a fully tractable form. The complete deformation procedure then yields two new physically viable anisotropic solutions, passing all standard stability and energy condition tests. Our results show, for the first time, that the direction of energy transfer between the seed sector and the decoupled source is uniquely governed by the deformation parameter, providing direct physical insight into the coupling between known and generic gravitational fields. Furthermore, we find that anisotropy plays a decisive role in the stability criteria of these stars, highlighting its nontrivial influence on realistic stellar evolution. These results offer a new perspective on the modeling of high-density stellar interiors and open a pathway for extending gravitational decoupling analyses to more complex astrophysical scenarios.
|
[
{
"version": "v1",
"created": "Tue, 01 Oct 2024 04:15:21 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 01:55:48 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Yousaf",
"Z.",
""
],
[
"Bamba",
"Kazuharu",
""
],
[
"Bhatti",
"M. Z.",
""
],
[
"Khan",
"S.",
""
]
] |
2410.03290
|
Haibo Wang
|
Haibo Wang, Zhiyang Xu, Yu Cheng, Shizhe Diao, Yufan Zhou, Yixin Cao, Qifan Wang, Weifeng Ge, Lifu Huang
|
Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language Models
|
Accepted by EMNLP 2025 Findings
| null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Video Large Language Models (Video-LLMs) have demonstrated remarkable capabilities in coarse-grained video understanding, however, they struggle with fine-grained temporal grounding. In this paper, we introduce Grounded-VideoLLM, a novel Video-LLM adept at perceiving and reasoning over specific video moments in a fine-grained manner. We identify that current Video-LLMs have limitations for fine-grained video understanding since they lack effective temporal modeling and timestamp representation. In light of this, we sharpen our model by incorporating (1) an additional temporal stream to encode the relationships between frames and (2) discrete temporal tokens enriched with specific time knowledge to represent timestamps. To optimize the training of Grounded-VideoLLM, we employ a multi-stage training scheme, beginning with simple video-captioning tasks and progressively introducing video temporal grounding tasks of increasing complexity. To further enhance Grounded-VideoLLM's temporal reasoning capability, we also curate a grounded VideoQA dataset by an automatic annotation pipeline. Extensive experiments demonstrate that Grounded-VideoLLM not only excels in fine-grained grounding tasks such as temporal sentence grounding, dense video captioning, and grounded VideoQA, but also shows great potential as a versatile video assistant for general video understanding.
|
[
{
"version": "v1",
"created": "Fri, 04 Oct 2024 10:04:37 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:15:19 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Haibo",
""
],
[
"Xu",
"Zhiyang",
""
],
[
"Cheng",
"Yu",
""
],
[
"Diao",
"Shizhe",
""
],
[
"Zhou",
"Yufan",
""
],
[
"Cao",
"Yixin",
""
],
[
"Wang",
"Qifan",
""
],
[
"Ge",
"Weifeng",
""
],
[
"Huang",
"Lifu",
""
]
] |
2410.03730
|
Mehdi Ali
|
Mehdi Ali, Michael Fromm, Klaudia Thellmann, Jan Ebert, Alexander Arno Weber, Richard Rutmann, Charvi Jain, Max L\"ubbering, Daniel Steinigen, Johannes Leveling, Katrin Klug, Jasper Schulze Buschhoff, Lena Jurkschat, Hammam Abdelwahab, Benny J\"org Stein, Karl-Heinz Sylla, Pavel Denisov, Nicolo' Brandizzi, Qasid Saleem, Anirban Bhowmick, Lennard Helmer, Chelsea John, Pedro Ortiz Suarez, Malte Ostendorff, Alex Jude, Lalith Manjunath, Samuel Weinbach, Carolin Penke, Oleg Filatov, Fabio Barth, Paramita Mirza, Lucas Weber, Ines Wendler, Rafet Sifa, Fabian K\"uch, Andreas Herten, Ren\'e J\"akel, Georg Rehm, Stefan Kesselheim, Joachim K\"ohler, Nicolas Flores-Herr
|
Teuken-7B-Base & Teuken-7B-Instruct: Towards European LLMs
| null | null | null | null |
cs.CL cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
We present two multilingual LLMs, Teuken 7B-base and Teuken 7B-instruct, designed to embrace Europe's linguistic diversity by supporting all 24 official languages of the European Union. Trained on a dataset comprising around 60% non-English data and utilizing a custom multilingual tokenizer, our models address the limitations of existing LLMs that predominantly focus on English or a few high-resource languages. We detail the models' development principles, i.e., data composition, tokenizer optimization, and training methodologies. The models demonstrate strong performance across multilingual benchmarks, as evidenced by their performance on European versions of ARC, HellaSwag, and TruthfulQA.
|
[
{
"version": "v1",
"created": "Mon, 30 Sep 2024 16:05:38 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Oct 2024 17:09:40 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 03:42:41 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ali",
"Mehdi",
""
],
[
"Fromm",
"Michael",
""
],
[
"Thellmann",
"Klaudia",
""
],
[
"Ebert",
"Jan",
""
],
[
"Weber",
"Alexander Arno",
""
],
[
"Rutmann",
"Richard",
""
],
[
"Jain",
"Charvi",
""
],
[
"Lübbering",
"Max",
""
],
[
"Steinigen",
"Daniel",
""
],
[
"Leveling",
"Johannes",
""
],
[
"Klug",
"Katrin",
""
],
[
"Buschhoff",
"Jasper Schulze",
""
],
[
"Jurkschat",
"Lena",
""
],
[
"Abdelwahab",
"Hammam",
""
],
[
"Stein",
"Benny Jörg",
""
],
[
"Sylla",
"Karl-Heinz",
""
],
[
"Denisov",
"Pavel",
""
],
[
"Brandizzi",
"Nicolo'",
""
],
[
"Saleem",
"Qasid",
""
],
[
"Bhowmick",
"Anirban",
""
],
[
"Helmer",
"Lennard",
""
],
[
"John",
"Chelsea",
""
],
[
"Suarez",
"Pedro Ortiz",
""
],
[
"Ostendorff",
"Malte",
""
],
[
"Jude",
"Alex",
""
],
[
"Manjunath",
"Lalith",
""
],
[
"Weinbach",
"Samuel",
""
],
[
"Penke",
"Carolin",
""
],
[
"Filatov",
"Oleg",
""
],
[
"Barth",
"Fabio",
""
],
[
"Mirza",
"Paramita",
""
],
[
"Weber",
"Lucas",
""
],
[
"Wendler",
"Ines",
""
],
[
"Sifa",
"Rafet",
""
],
[
"Küch",
"Fabian",
""
],
[
"Herten",
"Andreas",
""
],
[
"Jäkel",
"René",
""
],
[
"Rehm",
"Georg",
""
],
[
"Kesselheim",
"Stefan",
""
],
[
"Köhler",
"Joachim",
""
],
[
"Flores-Herr",
"Nicolas",
""
]
] |
2410.05549
|
Samuel S. Y. Wong
|
Xing Fan, Gerald Gabrielse, Peter W. Graham, Harikrishnan Ramani, Samuel S. Y. Wong, and Yawen Xiao
|
Highly Excited Electron Cyclotron for QCD Axion and Dark-Photon Detection
|
48 pages, 8 figures. Published version
|
Phys.Rev.D 111 (2025) 7, 075022
|
10.1103/PhysRevD.111.075022
| null |
hep-ph astro-ph.CO hep-ex physics.atom-ph quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We propose using highly excited cyclotron states of a trapped electron to detect meV axion and dark photon dark matter, marking a significant improvement over our previous proposal and demonstration [Phys. Rev. Lett. 129, 261801]. When the axion mass matches the cyclotron frequency $\omega_c$, the cyclotron state is resonantly excited, with a transition probability proportional to its initial quantum number, $n_c$. The sensitivity is enhanced by taking $n_c \sim 10^6 \left( \frac{0.1~\text{meV}}{\omega_c} \right)^2$. By optimizing key experimental parameters, we minimize the required averaging time for cyclotron detection to $t_{\text{ave}} \sim 10^{-6} $ seconds, permitting detection of such a highly excited state before its decay. An open-endcap trap design enables the external photon signal to be directed into the trap, rendering our background-free detector compatible with large focusing cavities, such as the BREAD proposal, while capitalizing on their strong magnetic fields. Furthermore, the axion conversion rate can be coherently enhanced by incorporating layers of dielectrics with alternating refractive indices within the cavity. Collectively, these optimizations enable us to probe the QCD axion parameter space from 0.1 meV to 2.3 meV (25-560 GHz), covering a substantial portion of the predicted post-inflationary QCD axion mass range. This sensitivity corresponds to probing the kinetic mixing parameter of the dark photon down to $\epsilon \approx 2 \times 10^{-16}$.
|
[
{
"version": "v1",
"created": "Mon, 07 Oct 2024 23:20:18 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 17:41:54 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fan",
"Xing",
""
],
[
"Gabrielse",
"Gerald",
""
],
[
"Graham",
"Peter W.",
""
],
[
"Ramani",
"Harikrishnan",
""
],
[
"Wong",
"Samuel S. Y.",
""
],
[
"Xiao",
"Yawen",
""
]
] |
2410.06154
|
Muhammad Jehanzeb Mirza
|
M. Jehanzeb Mirza, Mengjie Zhao, Zhuoyuan Mao, Sivan Doveh, Wei Lin, Paul Gavrikov, Michael Dorkenwald, Shiqi Yang, Saurav Jha, Hiromi Wakaki, Yuki Mitsufuji, Horst Possegger, Rogerio Feris, Leonid Karlinsky, James Glass
|
GLOV: Guided Large Language Models as Implicit Optimizers for Vision Language Models
|
Code: https://github.com/jmiemirza/GLOV
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we propose GLOV, which enables Large Language Models (LLMs) to act as implicit optimizers for Vision-Language Models (VLMs) to enhance downstream vision tasks. GLOV prompts an LLM with the downstream task description, querying it for suitable VLM prompts (e.g., for zero-shot classification with CLIP). These prompts are ranked according to their fitness for the downstream vision task. In each respective optimization step, the ranked prompts are fed as in-context examples (with their accuracies) to equip the LLM with the knowledge of the type of prompts preferred by the downstream VLM. Furthermore, we explicitly guide the LLM's generation at each optimization step by adding an offset vector -- calculated from the embedding differences between previous positive and negative solutions -- to the intermediate layer of the network for the next generation. This offset vector biases the LLM generation toward the type of language the downstream VLM prefers, resulting in enhanced performance on the downstream vision tasks. We comprehensively evaluate our GLOV on two tasks: object recognition and the critical task of enhancing VLM safety. Our GLOV shows performance improvement by up to 15.0% and 57.5% for dual-encoder (e.g., CLIP) and encoder-decoder (e.g., LlaVA) models for object recognition and reduces the attack success rate (ASR) on state-of-the-art VLMs by up to $60.7\%$.
|
[
{
"version": "v1",
"created": "Tue, 08 Oct 2024 15:55:40 GMT"
},
{
"version": "v2",
"created": "Mon, 02 Dec 2024 14:49:55 GMT"
},
{
"version": "v3",
"created": "Wed, 08 Jan 2025 23:08:03 GMT"
},
{
"version": "v4",
"created": "Mon, 03 Feb 2025 21:37:14 GMT"
},
{
"version": "v5",
"created": "Wed, 05 Feb 2025 19:00:50 GMT"
},
{
"version": "v6",
"created": "Wed, 20 Aug 2025 19:02:24 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Mirza",
"M. Jehanzeb",
""
],
[
"Zhao",
"Mengjie",
""
],
[
"Mao",
"Zhuoyuan",
""
],
[
"Doveh",
"Sivan",
""
],
[
"Lin",
"Wei",
""
],
[
"Gavrikov",
"Paul",
""
],
[
"Dorkenwald",
"Michael",
""
],
[
"Yang",
"Shiqi",
""
],
[
"Jha",
"Saurav",
""
],
[
"Wakaki",
"Hiromi",
""
],
[
"Mitsufuji",
"Yuki",
""
],
[
"Possegger",
"Horst",
""
],
[
"Feris",
"Rogerio",
""
],
[
"Karlinsky",
"Leonid",
""
],
[
"Glass",
"James",
""
]
] |
2410.06507
|
Zhen-Xing Zhao
|
Yi-Peng Xing, Xiao-Qian Jiang, Zhen-Xing Zhao
|
Semileptonic $s\to u$ decays of singly heavy baryons
|
Updated version, almost completely rewritten
| null | null | null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
$s\to u$ weak decays of heavy flavor hadrons play a unique role in the field of heavy flavor physics. In this work, we investigate semileptonic $s\to u$ decays of singly heavy baryons in the light-front approach under the three-quark picture. Firstly, we provide all the form factors using this quark model, especially for $f_{3}$ and $g_{3}$, which are usually considered non-extractable in recent literature. We also discuss the heavy quark limit for the form factors, and some of our results differ from those in the literature. We find that, when considering the heavy quark limit, the $m_{s}$ correction may not be negligible. Secondly, we apply the obtained form factors to predict some observables in semileptonic decays. It is worth noting that, due to the extremely small phase space, lepton flavor universality (LFU) is sensitive to the masses of leptons, and precise measurement of LFU is expected to be an important tool for testing the standard model.
|
[
{
"version": "v1",
"created": "Wed, 09 Oct 2024 03:17:20 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 11:25:53 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Xing",
"Yi-Peng",
""
],
[
"Jiang",
"Xiao-Qian",
""
],
[
"Zhao",
"Zhen-Xing",
""
]
] |
2410.07329
|
Yifan Chen
|
Yifan Chen, Yuxiang Liu, Jing Shu, Bin Xu, Xiao Xue, Yanjie Zeng
|
Identifying the Quadrupolar Nature of Gravitational Wave Background through Space-based Missions
|
26 pages, 8 figures, accepted version in JCAP
|
JCAP08(2025)063
|
10.1088/1475-7516/2025/08/063
| null |
gr-qc astro-ph.CO astro-ph.HE astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
The stochastic gravitational wave background (SGWB) consists of an incoherent collection of waves from both astrophysical and cosmological sources. To distinguish the SGWB from noise, it is essential to verify its quadrupolar nature, exemplified by the cross-correlations among pairs of pulsars within a pulsar timing array, commonly referred to as the Hellings-Downs curve. We study how this quadrupolar signature manifests in correlations between general GW detector pairs, characterized by their antenna responses. Focusing on space-based missions including laser interferometers (LISA, Taiji, TianQin) and atom interferometers (AEDGE/MAGIS), we demonstrate how orbital motion dynamically modulates detector correlations through time-dependent separations and relative orientation shifts. These modulations encode unique statistical features that serve as definitive markers of the SGWB's quadrupolar nature. Our findings identify optimal configurations for these missions, offer forecasts for the time needed to identify the quadrupolar nature of the SGWB, and are applicable to both space-space and space-terrestrial correlations.
|
[
{
"version": "v1",
"created": "Wed, 09 Oct 2024 18:00:04 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Jul 2025 11:01:56 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Chen",
"Yifan",
""
],
[
"Liu",
"Yuxiang",
""
],
[
"Shu",
"Jing",
""
],
[
"Xu",
"Bin",
""
],
[
"Xue",
"Xiao",
""
],
[
"Zeng",
"Yanjie",
""
]
] |
2410.08704
|
Jinwon Lee
|
Evert W. Stolte, Jinwon Lee, Hester G. Vennema, Rik Broekhoven, Esther Teng, Allard J. Katan, Lukas M. Veldman, Philip Willke, Sander Otte
|
Single-shot readout of the nuclear spin of an on-surface atom
|
34 pages including 5 figures, 4 supplementary notes, 9 supplementary figures, and 1 supplementary table
|
Nature Communications 16, 7785 (2025)
|
10.1038/s41467-025-63232-5
| null |
cond-mat.mes-hall quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Nuclear spins owe their long-lived magnetic states to their excellent isolation from the environment. At the same time, a finite degree of interaction with their surroundings is necessary for reading and writing the spin state. Therefore, detailed knowledge of and control over the atomic environment of a nuclear spin is key to optimizing conditions for quantum information applications. While various platforms enabled single-shot readout of nuclear spins, their direct environments were either unknown or impossible to controllably modify on the atomic scale. Scanning tunnelling microscopy (STM), combined with electron spin resonance (ESR), provides atomic-scale information of individual nuclear spins via the hyperfine interaction. Here, we demonstrate single-shot readout of an individual $^{\text{49}}$Ti nuclear spin with an STM. Employing a pulsed measurement scheme, we find its lifetime to be in the order of seconds. Furthermore, we shed light on the pumping and relaxation mechanisms of the nuclear spin by investigating its response to both ESR driving and tunnelling current, which is supported by model calculations. These findings give an atomic-scale insight into the nature of nuclear spin relaxation and are relevant for the development of atomically assembled qubit platforms.
|
[
{
"version": "v1",
"created": "Fri, 11 Oct 2024 10:47:46 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Nov 2024 16:55:10 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 14:17:39 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Stolte",
"Evert W.",
""
],
[
"Lee",
"Jinwon",
""
],
[
"Vennema",
"Hester G.",
""
],
[
"Broekhoven",
"Rik",
""
],
[
"Teng",
"Esther",
""
],
[
"Katan",
"Allard J.",
""
],
[
"Veldman",
"Lukas M.",
""
],
[
"Willke",
"Philip",
""
],
[
"Otte",
"Sander",
""
]
] |
2410.09290
|
Gary Tom
|
Gary Tom, Stanley Lo, Samantha Corapi, Alan Aspuru-Guzik, Benjamin Sanchez-Lengeling
|
Ranking over Regression for Bayesian Optimization and Molecule Selection
|
14 + 4 pages, 5 + 3 figures
|
APL Machine Learning, Volume 3, pg. 036113 (2025)
|
10.1063/5.0272663
| null |
cs.LG cs.AI stat.ML
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Bayesian optimization (BO) has become an indispensable tool for autonomous decision-making across diverse applications from autonomous vehicle control to accelerated drug and materials discovery. With the growing interest in self-driving laboratories, BO of chemical systems is crucial for machine learning (ML) guided experimental planning. Typically, BO employs a regression surrogate model to predict the distribution of unseen parts of the search space. However, for the selection of molecules, picking the top candidates with respect to a distribution, the relative ordering of their properties may be more important than their exact values. In this paper, we introduce Rank-based Bayesian Optimization (RBO), which utilizes a ranking model as the surrogate. We present a comprehensive investigation of RBO's optimization performance compared to conventional BO on various chemical datasets. Our results demonstrate similar or improved optimization performance using ranking models, particularly for datasets with rough structure-property landscapes and activity cliffs. Furthermore, we observe a high correlation between the surrogate ranking ability and BO performance, and this ability is maintained even at early iterations of BO optimization when using ranking surrogate models. We conclude that RBO is an effective alternative to regression-based BO, especially for optimizing novel chemical compounds.
|
[
{
"version": "v1",
"created": "Fri, 11 Oct 2024 22:38:14 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Tom",
"Gary",
""
],
[
"Lo",
"Stanley",
""
],
[
"Corapi",
"Samantha",
""
],
[
"Aspuru-Guzik",
"Alan",
""
],
[
"Sanchez-Lengeling",
"Benjamin",
""
]
] |
2410.09806
|
Majid Hashemi
|
Majid Hashemi and Omid Seify
|
Observability of 2HDM charged Higgs boson in a misaligned scenario
|
21 pages, 24 figures, version to appear in PRD
|
Phys. Rev. D 112, 035029 (2025)
|
10.1103/rp5d-c795
| null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The large hadron collider precision measurement of $h_{125}$ properties suggests that a misaligned two Higgs doublet model type 1 is preferred among other types of the model. In this analysis, a lepton collider potential for charged Higgs boson observation is evaluated and compared with HL-LHC expectations. We focus on $H^{\pm} \to W^{\pm}h_{125}$ decay which is exclusively activated in misalignment scenario. An event analysis including fast detector simulation is performed and it is shown that some parts of the parameter space which are out of HL-LHC reach can be probed by a high energy lepton collider with signal significance exceeding $2\sigma$ and $5\sigma$ for exclusion or discovery. The region with highest sensitivity is in the middle range charged Higgs boson mass, i.e., $350 < m_{H^{\pm}} <450$ GeV and $\tan \beta \gtrsim4$ at non-zero $\cos(\beta-\alpha)$ values.
|
[
{
"version": "v1",
"created": "Sun, 13 Oct 2024 11:47:26 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Jul 2025 08:16:19 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Hashemi",
"Majid",
""
],
[
"Seify",
"Omid",
""
]
] |
2410.10633
|
Kate Finucane
|
Kate Finucane, Lorraine Brennan, Roberta De Vito, Massimiliano Russo, Isobel Claire Gormley
|
Missing data imputation using a truncated Gaussian infinite factor model with application to metabolomics data
| null | null | null | null |
stat.ME
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Metabolomics is the study of small molecules in biological samples. Metabolomics data are typically high-dimensional and contain highly correlated variables and frequent missing values. Both missing at random (MAR) data, due to acquisition or processing errors, and missing not at random (MNAR) data, caused by values falling below detection thresholds, are common. Thus, imputation is a critical component of downstream analysis. Existing imputation methods generally assume one type of data missingness mechanism, or impute values outside the data's physical constraints. A novel truncated Gaussian infinite factor analysis (TGIFA) model is proposed to perform statistically principled and physically realistic imputation in metabolomics data. By incorporating truncated Gaussian assumptions, TGIFA respects the data's physical constraints, while leveraging an infinite latent factor framework to capture high-dimensional dependencies without pre-specifying the number of latent factors. Our Bayesian inference approach enables uncertainty quantification in both the values of the imputed data, and the missing data mechanism. A computationally efficient exchange algorithm enables scalable posterior inference via Markov Chain Monte Carlo. We validate TGIFA through a comprehensive simulation study and demonstrate its utility in a motivating urinary metabolomics dataset, where it yields useful imputations, with associated uncertainty quantification. Open-source R code, available at https://github.com/kfinucane/TGIFA, accompanies TGIFA.
|
[
{
"version": "v1",
"created": "Mon, 14 Oct 2024 15:44:41 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 17:50:21 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Finucane",
"Kate",
""
],
[
"Brennan",
"Lorraine",
""
],
[
"De Vito",
"Roberta",
""
],
[
"Russo",
"Massimiliano",
""
],
[
"Gormley",
"Isobel Claire",
""
]
] |
2410.10752
|
Nicolas Fintzi
|
Nicolas Fintzi and Jean-Lou Pierson
|
Averaged equations for disperse two-phase flow with interfacial properties and their closures for dilute suspension of droplets
| null | null | null | null |
physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
This article provides a derivation of the averaged equations governing the motion of dispersed two-phase flows with interfacial transport. We begin by revisiting the two-fluid formulation, as well as the distributional form of the interfacial transport equation which holds on the entire domain. Following this, a general Lagrangian model is introduced, which accounts for the effects of both internal and interfacial properties of the dispersed inclusions (bubbles, droplets, or particles) within a continuous phase. This is achieved by derivation of conservation laws for particle surface and volume-integrated properties. By summing the internal and interfacial conservation laws, we derive a conservation equation for an arbitrary Lagrangian property associated with the inclusion. We then proceed by deriving the lesser-known conservation equations for the moments of the volume and surface distribution of an arbitrary Lagrangian property. Next, the averaged equations for the dispersed phase are derived through two distinct approaches: the particle-averaged (or Lagrangian-based) formalism, and the phase-averaged method. One important conclusion of this work is the demonstration of the relationship between the particle-averaged and phase-averaged equations. We show that the dispersed phase-averaged equations can be interpreted as a series expansion of the particle-averaged moment equations. We then present a "hybrid" set of equations, consisting of phase-averaged equations for the continuous fluid phase, complemented by an arbitrary number of moment conservation.
|
[
{
"version": "v1",
"created": "Mon, 14 Oct 2024 17:24:52 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:11:47 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Fintzi",
"Nicolas",
""
],
[
"Pierson",
"Jean-Lou",
""
]
] |
2410.11638
|
Ilya Chevyrev
|
Ilya Chevyrev and Hora Mirsajjadi
|
Local well-posedness of subcritical non-linear heat equations with Gaussian initial data
|
45 pages, 3 Figures. Minor revisions. Accepted version in Journal of Functional Analysis
|
Journal of Functional Analysis, Volume 289, Issue 12 (2025), 111160
|
10.1016/j.jfa.2025.111160
| null |
math.PR math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
We show that any non-linear heat equation with scaling critical dimension $-1$ is locally well-posed when its initial condition is taken as the Gaussian free field in fractional dimension $d < 4$. Our results in particular extend the well-posedness results of arXiv:2111.10652, arXiv:2201.03487 from $d=3$ to the entire subcritical regime.
|
[
{
"version": "v1",
"created": "Tue, 15 Oct 2024 14:29:14 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 10:29:35 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Chevyrev",
"Ilya",
""
],
[
"Mirsajjadi",
"Hora",
""
]
] |
2410.14614
|
Ali K. Shargh
|
A. K. Shargh, C. D. Stiles, J. A. El-Awady
|
Temperature-dependent discovery of BCC refractory multi-principal element alloys: Integrating deep learning and CALPHAD calculations
| null | null | null | null |
cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
Single-phase body-centered cubic (BCC) refractory multi-principal element alloys (RMPEAs) offer potential for developing alloys with exceptional strength. However, the compositional design space is immense. Exhaustively mapping this space with conventional CALculation of PHAse Diagrams (CALPHAD) is impractical because database coverage and run times scale poorly with millions of candidate chemistries. To address this, we train a deep-learning surrogate on CALPHAD outputs that preserves the thermodynamic fidelity while accelerating temperature-dependent phase-fraction predictions of RMPEA phases. The model achieves high accuracy in predicting phase fractions for up to eight distinct phases across different temperatures and offers a speedup of two orders of magnitude compared to CALPHAD. Using this model, we screen the Ti, Fe, Al, V, Ni, Nb and Zr elemental space for potentially stable single-phase BCC alloys at different annealing temperatures and extract design insights to guide the synthesis of new BCC RMPEAs in experiments. Finally, we develop an analytical model that enables rapid, interpretable identification of single-phase BCC RMPEAs.
|
[
{
"version": "v1",
"created": "Fri, 18 Oct 2024 17:11:33 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 16:14:56 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Shargh",
"A. K.",
""
],
[
"Stiles",
"C. D.",
""
],
[
"El-Awady",
"J. A.",
""
]
] |
2410.14906
|
Eleftherios Ioannidis
|
Eleftherios Ioannidis, Yannick Zakowski, Steve Zdancewic, Sebastian Angel
|
Structural Temporal Logic for Mechanized Program Verification
| null |
Proc. ACM Program. Lang., Vol. 9, No. OOPSLA2, Article 313. Publication date: October 2025
|
10.1145/3763091
| null |
cs.PL cs.LO
|
http://creativecommons.org/licenses/by/4.0/
|
Mechanized verification of liveness properties for infinite programs with effects and nondeterminism is challenging. Existing temporal reasoning frameworks operate at the level of models such as traces and automata. Reasoning happens at a very low-level, requiring complex nested (co-)inductive proof techniques and familiarity with proof assistant mechanics (e.g., the guardedness checker). Further, reasoning at the level of models instead of program constructs creates a verification gap that loses the benefits of modularity and composition enjoyed by structural program logics such as Hoare Logic. To address this verification gap, and the lack of compositional proof techniques for temporal specifications, we propose Ticl, a new structural temporal logic. Using ticl, we encode complex (co-)inductive proof techniques as structural lemmas and focus our reasoning on variants and invariants. We show that it is possible to perform compositional proofs of general temporal properties in a proof assistant, while working at a high level of abstraction. We demonstrate the benefits of Ticl by giving mechanized proofs of safety and liveness properties for programs with scheduling, concurrent shared memory, and distributed consensus, demonstrating a low proof-to-code ratio.
|
[
{
"version": "v1",
"created": "Fri, 18 Oct 2024 23:19:29 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Oct 2024 18:11:15 GMT"
},
{
"version": "v3",
"created": "Mon, 18 Nov 2024 04:46:20 GMT"
},
{
"version": "v4",
"created": "Sat, 08 Feb 2025 17:01:51 GMT"
},
{
"version": "v5",
"created": "Tue, 25 Mar 2025 17:53:49 GMT"
},
{
"version": "v6",
"created": "Mon, 07 Apr 2025 18:21:31 GMT"
},
{
"version": "v7",
"created": "Thu, 21 Aug 2025 16:35:36 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Ioannidis",
"Eleftherios",
""
],
[
"Zakowski",
"Yannick",
""
],
[
"Zdancewic",
"Steve",
""
],
[
"Angel",
"Sebastian",
""
]
] |
2410.15186
|
Adam Kiehl
|
Mayla R. Boguslav, Adam Kiehl, David Kott, G. Joseph Strecker, Tracy Webb, Nadia Saklou, Terri Ward, Michael Kirby
|
Fine-tuning foundational models to code diagnoses from veterinary health records
|
26 pages, 5 figures
| null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Veterinary medical records represent a large data resource for application to veterinary and One Health clinical research efforts. Use of the data is limited by interoperability challenges including inconsistent data formats and data siloing. Clinical coding using standardized medical terminologies enhances the quality of medical records and facilitates their interoperability with veterinary and human health records from other sites. Previous studies, such as DeepTag and VetTag, evaluated the application of Natural Language Processing (NLP) to automate veterinary diagnosis coding, employing long short-term memory (LSTM) and transformer models to infer a subset of Systemized Nomenclature of Medicine - Clinical Terms (SNOMED-CT) diagnosis codes from free-text clinical notes. This study expands on these efforts by incorporating all 7,739 distinct SNOMED-CT diagnosis codes recognized by the Colorado State University (CSU) Veterinary Teaching Hospital (VTH) and by leveraging the increasing availability of pre-trained language models (LMs). 13 freely-available pre-trained LMs were fine-tuned on the free-text notes from 246,473 manually-coded veterinary patient visits included in the CSU VTH's electronic health records (EHRs), which resulted in superior performance relative to previous efforts. The most accurate results were obtained when expansive labeled data were used to fine-tune relatively large clinical LMs, but the study also showed that comparable results can be obtained using more limited resources and non-clinical LMs. The results of this study contribute to the improvement of the quality of veterinary EHRs by investigating accessible methods for automated coding and support both animal and human health research by paving the way for more integrated and comprehensive health databases that span species and institutions.
|
[
{
"version": "v1",
"created": "Sat, 19 Oct 2024 19:30:29 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 23:47:43 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Boguslav",
"Mayla R.",
""
],
[
"Kiehl",
"Adam",
""
],
[
"Kott",
"David",
""
],
[
"Strecker",
"G. Joseph",
""
],
[
"Webb",
"Tracy",
""
],
[
"Saklou",
"Nadia",
""
],
[
"Ward",
"Terri",
""
],
[
"Kirby",
"Michael",
""
]
] |
2410.15885
|
Zj Tang
|
Zuojin Tang, Bin Hu, Chenyang Zhao, De Ma, Gang Pan, Bin Liu
|
VLASCD: A Visual Language Action Model for Simultaneous Chatting and Decision Making
|
Accepted by EMNLP 2025 (Main Conference)
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Although current mainstream pre-trained large models, such as LLM models represented by ChatGPT and VLA models represented by OpenVLA, have achieved significant progress in multimodal tasks through a "Multiple-Input, Single-Output" (MISO) architecture. However, our investigation reveals that the MISO architecture exhibits fundamental limitations in "Multiple-Input, Multiple-Output" (MIMO) (e.g., parallel multi-tasks output processing): the architecture generates task mutual exclusion effects, leading to resource contention among different tasks when sharing output channels, and consequently resulting in optimization imbalance and performance degradation. In contrast, human MIMO processing inherently enables concurrent task execution (e.g., while dialogue and decision-making) without interference. Inspired by this, in this work, we propose a unified MIMO training model with parallel multi-tasks output capabilities termed Visual Language Action Model for Simultaneously Chatting and Decision Making. We refer to this method as VLASCD or MIMO-VLA, and in the following, we will use these two names interchangeably. We evaluate the model on the CARLA autonomous driving platform. The results show that, compared to LLM models with MISO dialogue capabilities, reinforcement learning models, and VLA models with MISO decision-making capabilities, MIMO-VLA significantly outperforms existing MISO models in simultaneously handling dialogue generation and decision-making tasks within the MIMO scenario.
|
[
{
"version": "v1",
"created": "Mon, 21 Oct 2024 11:02:42 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 02:12:56 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Tang",
"Zuojin",
""
],
[
"Hu",
"Bin",
""
],
[
"Zhao",
"Chenyang",
""
],
[
"Ma",
"De",
""
],
[
"Pan",
"Gang",
""
],
[
"Liu",
"Bin",
""
]
] |
2410.16611
|
Yijie Huang
|
Lijun Bo, Yijie Huang, Kaixin Yan, Xiang Yu
|
Optimal consumption under relaxed benchmark tracking and consumption drawdown constraint
| null | null | null | null |
math.OC q-fin.MF
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper studies an optimal consumption problem with both relaxed benchmark tracking and consumption drawdown constraint, leading to a stochastic control problem with dynamic state-control constraints. In our relaxed tracking formulation, it is assumed that the fund manager can strategically inject capital to the fund account such that the total capital process always outperforms the benchmark process, which is described by a geometric Brownian motion. We first transform the original regular-singular control problem with state-control constraints into an equivalent regular control problem with a reflected state process and consumption drawdown constraint. By utilizing the dual transform and the optimal consumption behavior, we then turn to study the linear dual PDE with both Neumann boundary condition and free boundary condition in a piecewise manner across different regions. Using the smoothfit principle and the super-contact condition, we derive the closed-form solution of the dual PDE, and obtain the optimal investment and consumption in feedback form. We then prove the verification theorem on optimality by some novel arguments with the aid of an auxiliary reflected dual process and some technical estimations. Some numerical examples and financial insights are also presented.
|
[
{
"version": "v1",
"created": "Tue, 22 Oct 2024 01:49:25 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:41:00 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 01:12:29 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Bo",
"Lijun",
""
],
[
"Huang",
"Yijie",
""
],
[
"Yan",
"Kaixin",
""
],
[
"Yu",
"Xiang",
""
]
] |
2410.16671
|
Jiamu Wang
|
Jiamu Wang, Jin Tae Kwak
|
NucleiMix: Realistic Data Augmentation for Nuclei Instance Segmentation
| null | null | null | null |
eess.IV cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Nuclei instance segmentation is an essential task in pathology image analysis, serving as the foundation for many downstream applications. The release of several public datasets has significantly advanced research in this area, yet many existing methods struggle with data imbalance issues. To address this challenge, this study introduces a data augmentation method, called NucleiMix, which is designed to balance the distribution of nuclei types by increasing the number of rare-type nuclei within datasets. NucleiMix operates in two phases. In the first phase, it identifies candidate locations similar to the surroundings of rare-type nuclei and inserts rare-type nuclei into the candidate locations. In the second phase, it employs a progressive inpainting strategy using a pre-trained diffusion model to seamlessly integrate rare-type nuclei into their new environments in replacement of major-type nuclei or background locations. We systematically evaluate the effectiveness of NucleiMix on three public datasets using two popular nuclei instance segmentation models. The results demonstrate the superior ability of NucleiMix to synthesize realistic rare-type nuclei and to enhance the quality of nuclei segmentation and classification in an accurate and robust manner.
|
[
{
"version": "v1",
"created": "Tue, 22 Oct 2024 04:03:36 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 05:47:34 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Wang",
"Jiamu",
""
],
[
"Kwak",
"Jin Tae",
""
]
] |
2410.21162
|
Ryoto Iwai
|
A. Antognini, N. J. Ayres, I. Belosevic, V. Bondar, A. Eggenberger, M. Hildebrandt, R. Iwai, K. Kirch, A. Knecht, G. Lospalluto, J. Nuber, A. Papa, M. Sakurai, I. Solovyev, D. Taqqu, T. Yan
|
Phase space compression of a positive muon beam in two spatial dimensions
|
Submission to SciPost
| null | null | null |
physics.acc-ph hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
We present the first demonstration of simultaneous phase space compression in two spatial dimensions of a positive muon beam, the first stage of the novel high-brightness muon beam under development by the muCool collaboration at the Paul Scherrer Institute. The keV-energy, sub-mm size beam would enable a factor 10$^5$ improvement in brightness for precision muSR, and atomic and particle physics measurements with positive muons. This compression is achieved within a cryogenic helium gas target with a strong density gradient, placed in a homogeneous magnetic field, under the influence of a complex electric field. In the next phase, the muon beam will be extracted into vacuum.
|
[
{
"version": "v1",
"created": "Mon, 28 Oct 2024 16:03:41 GMT"
},
{
"version": "v2",
"created": "Thu, 21 Aug 2025 14:17:17 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Antognini",
"A.",
""
],
[
"Ayres",
"N. J.",
""
],
[
"Belosevic",
"I.",
""
],
[
"Bondar",
"V.",
""
],
[
"Eggenberger",
"A.",
""
],
[
"Hildebrandt",
"M.",
""
],
[
"Iwai",
"R.",
""
],
[
"Kirch",
"K.",
""
],
[
"Knecht",
"A.",
""
],
[
"Lospalluto",
"G.",
""
],
[
"Nuber",
"J.",
""
],
[
"Papa",
"A.",
""
],
[
"Sakurai",
"M.",
""
],
[
"Solovyev",
"I.",
""
],
[
"Taqqu",
"D.",
""
],
[
"Yan",
"T.",
""
]
] |
2410.21673
|
Peng Liang
|
Lin Li, Xinchun Yu, Xinyu Chen, Peng Liang
|
Knowledge-Guided Prompt Learning for Request Quality Assurance in Public Code Review
|
27 pages, 5 images, 12 tables, Manuscript revision submitted to a journal (2025)
| null | null | null |
cs.SE cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Public Code Review (PCR) is developed in the Software Question Answering (SQA) community, assisting developers in exploring high-quality and efficient review services. Current methods on PCR mainly focus on the reviewer's perspective, including finding a capable reviewer, predicting comment quality, and recommending/generating review comments. However, it is not well studied that how to satisfy the review necessity requests posted by developers which can increase their visibility, which in turn acts as a prerequisite for better review responses. To this end, we propose K nowledge-guided P rompt learning for P ublic Code Review (KP-PCR) to achieve developer-based code review request quality assurance (i.e., predicting request necessity and recommending tags subtask). Specifically, we reformulate the two subtasks via 1) text prompt tuning which converts both of them into a Masked Language Model (MLM) by constructing prompt templates using hard prompt; and 2) knowledge and code prefix tuning which introduces knowledge guidance from fine-tuned large language models by soft prompt, and uses program dependence graph to characterize code snippets. Finally, both of the request necessity prediction and tag recommendation subtasks output predicted results through an answer engineering module. In addition, we further analysis the time complexity of our KP-PCR that has lightweight prefix based the operation of introducing knowledge guidance. Experimental results on the PCR dataset for the period 2011-2023 demonstrate that our KP-PCR outperforms baselines by 2.3%-8.4% in the request necessity prediction and by 1.4%-6.9% in the tag recommendation. The code implementation is released at https://github.com/WUT-IDEA/KP-PCR.
|
[
{
"version": "v1",
"created": "Tue, 29 Oct 2024 02:48:41 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 11:43:05 GMT"
},
{
"version": "v3",
"created": "Thu, 21 Aug 2025 17:33:04 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Li",
"Lin",
""
],
[
"Yu",
"Xinchun",
""
],
[
"Chen",
"Xinyu",
""
],
[
"Liang",
"Peng",
""
]
] |
2410.21751
|
Ran Cheng
|
Junyu Tang, Hantao Zhang, Ran Cheng
|
N\'eel Spin-Orbit Torque in Antiferromagnetic Quantum Spin and Anomalous Hall Insulators
| null |
Nature Communications 16, 7790 (2025)
|
10.1038/s41467-025-63171-1
| null |
cond-mat.mes-hall cond-mat.mtrl-sci quant-ph
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Interplay between magnetic ordering and topological electrons not only enables new topological phases but also underpins electrical control of magnetism. Here we extend the Kane-Mele model to include the exchange coupling to a collinear background antiferromagnetic (AFM) order, which can describe transition metal trichalcogenides. Owing to the spin-orbit coupling and staggered on-site potential, the system could exhibit the quantum anomalous Hall and quantum spin Hall effects in the absence of a net magnetization. Besides the chiral edge states, these topological phases support a staggered Edelstein effect through which an applied electric field can generate opposite non-equilibrium spins on the two AFM sublattices, realizing the N\'eel-type spin-orbit torque (NSOT). Contrary to known NSOTs in AFM metals driven by conduction currents, our NSOT arises from pure adiabatic currents devoid of Joule heating, while being a bulk effect not carried by the edge currents. By virtue of the NSOT, the electric field of a microwave can drive the AFM dynamics with a remarkably high efficiency. Compared to the ordinary AFM resonance driven by the magnetic field, the new mechanism can enhance the resonance amplitude by more than one order of magnitude and the absorption rate of the microwave power by over two orders of magnitude. Our findings unravel an incredible way to exploit AFM topological phases to achieve ultrafast magnetic dynamics.
|
[
{
"version": "v1",
"created": "Tue, 29 Oct 2024 05:36:56 GMT"
},
{
"version": "v2",
"created": "Mon, 21 Jul 2025 19:16:28 GMT"
}
] | 2025-08-22T00:00:00 |
[
[
"Tang",
"Junyu",
""
],
[
"Zhang",
"Hantao",
""
],
[
"Cheng",
"Ran",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.