From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models Paper • 2607.06553 • Published 27 days ago • 20
SPACENUM: Revisiting Spatial Numerical Understanding in VLMs Paper • 2605.23898 • Published May 22 • 7
Reward-Decomposed Reinforcement Learning for Immersive Video Role-Playing Paper • 2605.04733 • Published Jun 3
On-Policy Distillation with Best-of-N Teacher Rollout Selection Paper • 2605.09725 • Published May 13
ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model Paper • 2603.22281 • Published Mar 23 • 20
Vision Language Models Cannot Reason About Physical Transformation Paper • 2603.07109 • Published Mar 7 • 2
SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks Paper • 2602.12670 • Published Feb 13 • 65
Less is Enough: Synthesizing Diverse Data in Feature Space of LLMs Paper • 2602.10388 • Published Feb 11 • 246
BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers? Paper • 2510.18003 • Published Oct 20, 2025
Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty? Paper • 2605.12684 • Published May 12 • 11
CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards Paper • 2510.08529 • Published Oct 9, 2025 • 19