Proxy Exploration and Reusable Guidance: A Modular LLM Post-Training Paradigm via Proxy-Guided Update Signals Paper • 2607.11505 • Published 4 days ago • 13
COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation Paper • 2605.31264 • Published May 29 • 124
All Roads Lead to Rome: Incentivizing Divergent Thinking in Vision-Language Models Paper • 2604.00479 • Published Apr 1 • 70
Few Tokens Matter: Entropy Guided Attacks on Vision-Language Models Paper • 2512.21815 • Published Dec 26, 2025 • 22
More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models Paper • 2509.25848 • Published Sep 30, 2025 • 81