Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability
Paper • 2604.06628 • Published • 189
Qinghua Mao
maopopovich
·
AI & ML interests
None yet
Recent Activity
upvoted a paper 3 days ago
ATBench: A Diverse and Realistic Trajectory Benchmark for Long-Horizon Agent Safety upvoted a paper about 1 month ago
Code2Math: Can Your Code Agent Effectively Evolve Math Problems Through Exploration?Organizations
upvoted a paper 2 days ago
upvoted a paper 3 days ago
upvoted a paper about 1 month ago
upvoted an article about 1 year ago
Article
What is test-time compute and how to scale it?
•
117