{
  "id": 1489,
  "url": "https://arxiv.org/abs/2606.05025v1",
  "title": "Invariant Gradient Alignment for Robust Reasoning Distillation",
  "summary": "Large language models (LLMs) suffer from shortcut learning: they systematically fail on out-of-distribution (OOD) inputs whose semantic surface differs from training data, even when the logical structure is identical. This undermines knowledge distillation pipelines that transfer chain-of-thought reasoning to smaller students. We introduce Invariant Gradient Alignment (IGA), a training framework that aligns gradient updates across semantically diverse but logically isomorphic examples via three ",
  "authors": "Zehua Cheng, Wei Dai, Jiahao Sun",
  "category": "research",
  "topics": "safety-alignment,children-education",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-03T15:48:52.000Z",
  "fetched_at": "2026-07-14T14:15:17.104Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/1489",
  "original_url": "https://arxiv.org/abs/2606.05025v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}