APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies Paper • 2606.12366 • Published 16 days ago • 5
Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs Paper • 2605.30611 • Published 29 days ago • 247
Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players Paper • 2605.28816 • Published 30 days ago • 431
SpaceDG: Benchmarking Spatial Intelligence under Visual Degradation Paper • 2605.22536 • Published May 21 • 28
Conditional Equivalence of DPO and RLHF: Implicit Assumption, Failure Modes, and Provable Alignment Paper • 2605.20834 • Published May 20 • 5
SQuTR: A Robustness Benchmark for Spoken Query to Text Retrieval under Acoustic Noise Paper • 2602.12783 • Published Feb 13 • 246
APEX: Large-scale Multi-task Aesthetic-Informed Popularity Prediction for AI-Generated Music Paper • 2605.03395 • Published May 5 • 6