The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes
Paper • 2605.11182 • Published • 4
Shengxuan Qiu
SXQiu
AI & ML interests
Efficient AI, LLM Reasoning
Recent Activity
upvoted a paper about 23 hours ago
The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes upvoted a paper about 2 months ago
AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents authored a paper about 2 months ago
HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and ReductionOrganizations
None yet