tzjz89

tzjz89

8 4

·

AI & ML interests

NLP

Organizations

upvoted a collection 6 months ago

Deepseek Papers

Deepseek papers collection • 32 items • Updated 6 days ago • 360

upvoted a paper 8 months ago

Stabilizing Reinforcement Learning with LLMs: Formulation and Practices

Paper • 2512.01374 • Published Dec 1, 2025 • 107

upvoted 3 papers about 1 year ago

Group Sequence Policy Optimization

Paper • 2507.18071 • Published Jul 24, 2025 • 320

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

Paper • 2506.01939 • Published Jun 2, 2025 • 190

Understanding R1-Zero-Like Training: A Critical Perspective

Paper • 2503.20783 • Published Mar 26, 2025 • 60

upvoted 2 papers over 1 year ago

Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Models

Paper • 2501.11873 • Published Jan 21, 2025 • 68

ProcessBench: Identifying Process Errors in Mathematical Reasoning

Paper • 2412.06559 • Published Dec 9, 2024 • 87

upvoted a collection over 2 years ago

Qwen1.5

Qwen1.5 is the improved version of Qwen, the large language model series developed by Alibaba Cloud. • 53 items • Updated Mar 2 • 214