Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning Paper • 2506.01939 • Published Jun 2 • 185
Understanding R1-Zero-Like Training: A Critical Perspective Paper • 2503.20783 • Published Mar 26 • 56
Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Models Paper • 2501.11873 • Published Jan 21 • 66
ProcessBench: Identifying Process Errors in Mathematical Reasoning Paper • 2412.06559 • Published Dec 9, 2024 • 84
Qwen1.5 Collection Qwen1.5 is the improved version of Qwen, the large language model series developed by Alibaba Cloud. • 55 items • Updated Jul 21 • 210
