Rethinking On-Policy Self-Distillation for Thinking Models
Paper • 2607.05184 • Published • 2
Simran Kaur
kaur-sim
·
AI & ML interests
None yet
Recent Activity
authored a paper 26 days ago
Skill-Mix: a Flexible and Expandable Family of Evaluations for AI models authored a paper 26 days ago
Instruct-SkillMix: A Powerful Pipeline for LLM Instruction Tuning authored a paper 26 days ago
Can Models Learn Skill Composition from Examples?Organizations
upvoted a paper 28 days ago