Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs Paper • 2605.30611 • Published 7 days ago • 170
Speculative Pipeline Decoding: Higher-Accruacy and Zero-Bubble Speculation via Pipeline Parallelism Paper • 2605.30852 • Published 6 days ago • 9
VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization Paper • 2606.02564 • Published 3 days ago • 25
Forecasting Downstream Performance of LLMs With Proxy Metrics Paper • 2605.18607 • Published 17 days ago • 14
IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools Paper • 2605.20682 • Published 15 days ago • 83
Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics Paper • 2605.18549 • Published 17 days ago • 2
SkillsVote: Lifecycle Governance of Agent Skills from Collection, Recommendation to Evolution Paper • 2605.18401 • Published 17 days ago • 126
RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time Paper • 2604.11626 • Published Apr 13 • 102
LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models Paper • 2603.28301 • Published Mar 30 • 82
Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability Paper • 2604.06628 • Published Apr 8 • 326
Adam's Law: Textual Frequency Law on Large Language Models Paper • 2604.02176 • Published Apr 2 • 504
CLEAR: Unlocking Generative Potential for Degraded Image Understanding in Unified Multimodal Models Paper • 2604.04780 • Published Apr 6 • 11