{
  "id": 1180,
  "url": "https://arxiv.org/abs/2606.11173v1",
  "title": "The Role of Feedback Alignment in Self-Distillation",
  "summary": "Conditioning a language model on additional context, such as feedback on a previous attempt, typically improves its response. Self-distillation trains the model to retain this improvement when the context is not present. The method works by matching the model's output distribution under two settings: a student that sees only the question, and a self-teacher that also sees the context. What the model learns therefore depends on what context the self-teacher receives, yet the design of this contex",
  "authors": "Semih Kara, Oğuzhan Ersoy",
  "category": "research",
  "topics": "safety-alignment,children-education",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-09T17:50:09.000Z",
  "fetched_at": "2026-07-14T14:15:03.617Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/1180",
  "original_url": "https://arxiv.org/abs/2606.11173v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}