arxiv:2605.17026
Nguyễn Minh Phúc
DatPySci
AI & ML interests
Reinforcement learning, NLP
Recent Activity
authored a paper 10 days ago
Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road authored a paper 10 days ago
The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models upvoted an article about 1 month ago
Navigating the RLHF Landscape: From Policy Gradients to PPO, GAE, and DPO for LLM AlignmentOrganizations
datasets 53
DatPySci/OpenThoughts-mixture-preprocessed
Viewer • Updated • 114k • 38
DatPySci/OpenThoughts-114k-preprocessed
Viewer • Updated • 114k • 38
DatPySci/Self-Distillation-Qwen2.5-7B-Instruct-medreason-prefix-instruct-tokenized
Viewer • Updated • 32.7k • 10
DatPySci/Self-Distillation-Qwen3-4B-Thinking-2507-medreason-prefix-instruct-tokenized
Viewer • Updated • 32.7k • 27
DatPySci/Self-Distillation-Qwen3-8B-medreason-prefix-instruct-tokenized
Viewer • Updated • 32.7k • 30
DatPySci/Self-Distillation-Qwen3-4B-Thinking-2507-medreason-tokenized
Viewer • Updated • 32.7k • 21
DatPySci/Self-Distillation-Qwen3-4B-Instruct-2507-medreason-tokenized
Viewer • Updated • 32.7k • 10
DatPySci/Self-Distillation-Qwen3-8B-medreason-tokenized
Viewer • Updated • 32.7k • 38
DatPySci/MedReason-tokenized
Viewer • Updated • 32.7k • 8
DatPySci/Self-Distillation-m23k-tokenized
Viewer • Updated • 23.5k • 9