DARE-bench: Evaluating Modeling and Instruction Fidelity of LLMs in Data Science Paper • 2602.24288 • Published 4 days ago • 2
Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning Paper • 2602.10090 • Published 21 days ago • 51
Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning Paper • 2506.01939 • Published Jun 2, 2025 • 188