Tag
The paper introduces Reflection-Enhanced Self-Distillation (Resd), a framework that transforms failure feedback into corrective supervision for LLMs, enabling efficient learning from rare successes. It outperforms standard self-distillation baselines and achieves faster early improvement than GRPO with fewer samples.
A user reports that Claude Code sometimes hangs for three minutes without providing status updates or feedback on whether it is still processing.
BugDrop is an in-app feedback tool that automatically creates GitHub Issues with screenshots, streamlining bug reporting for development teams.