Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information

arXiv ID: 2605.11609 HF Paper: 2605.11609 Source: HuggingFace Daily Papers Upvotes: 15 Submitter: yeates

---

Notes

• Ingested from HuggingFace Daily Papers via kb-ingest cron job (2026-05-20T09:09:21Z).
• Ranked by community upvotes.