{
  "id": 1579,
  "url": "https://arxiv.org/abs/2606.28186",
  "title": "Cognitive Episodes in LLM Reasoning Traces Enable Interpretable Human Item Difficulty Prediction",
  "summary": "arXiv:2606.28186v2 Announce Type: replace-cross Abstract: Predicting human item difficulty is central to educational assessment, where reliable estimates support fairness and effective test construction. Existing methods often depend on costly human calibration or item-level textual representations, providing limited evidence about the cognitive processes that make items difficult. We argue that difficulty should be viewed not only as a property of item text, but also as an observable consequenc",
  "authors": "Chenguang Wang, Ming Li, Xinyue Zeng, Zhuochun Li, Hong Jiao, Tianyi Zhou, Dawei Zhou",
  "category": "research",
  "topics": "bias-fairness",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-14T04:00:00.000Z",
  "fetched_at": "2026-07-14T16:04:12.223Z",
  "source_slug": "arxiv-cscy",
  "source_name": "arXiv cs.CY",
  "source_homepage": "https://arxiv.org/list/cs.CY/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/1579",
  "original_url": "https://arxiv.org/abs/2606.28186",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}