{"version":1,"type":"story","url":"https://digestai.news/story/researchers-present-citeanchor-framework-for-detecting-hallucinated-ci","json":"https://digestai.news/story/researchers-present-citeanchor-framework-for-detecting-hallucinated-ci.json","markdown":"https://digestai.news/story/researchers-present-citeanchor-framework-for-detecting-hallucinated-ci.md","slug":"researchers-present-citeanchor-framework-for-detecting-hallucinated-ci","headline":"Researchers present CiteAnchor framework for detecting hallucinated citations","summary":"A new benchmark called CiteResolve, containing 423 real‑world hallucinated citations annotated for source recoverability, is introduced alongside a source‑aware grounding system named CiteAnchor. The framework tailors retrieval and field‑level verification to each source type and separately judges citation validity and whether a unique source can be recovered.\\n\\nOn an existing detection benchmark, CiteAnchor reaches an F1 score of 0.841 when paired with GPT‑5.6 Luna and 0.784 with an open‑weight backbone, beating the same backbones by 7.9 and 36.8 points respectively. When evaluated on the CiteResolve benchmark, the end‑to‑end task success rates are 74.7% and 70.7%, outperforming comparable web‑search baselines by 16.3 and 65.3 percentage points. The authors have released the code and data on GitHub for further research.","keyPoints":["CiteAnchor framework achieves 0.841 F1 with GPT‑5.6 Luna on citation detection benchmark.","CiteResolve benchmark includes 423 annotated hallucinated citations for source recoverability.","End‑to‑end success on CiteResolve is 74.7% and 70.7%, beating web‑search baselines by up to 65.3 points."],"whyItMatters":"Improving detection and resolution of fabricated citations helps maintain scientific credibility and reduces the spread of misinformation in AI‑generated research.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["OpenAI"],"models":["GPT-5.6 Luna"],"people":[]},"firstPublishedAt":"2026-10-23T04:00:00Z","updatedAt":"2026-10-23T04:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"arXiv cs.CL","title":"CiteAnchor: Source-Aware Grounding for Hallucinated Citation Detection and Resolution","url":"https://arxiv.org/abs/2610.30355","publishedAt":"2026-10-23T04:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Researchers present CiteAnchor framework for detecting hallucinated citations\", 23 October 2026, https://digestai.news/story/researchers-present-citeanchor-framework-for-detecting-hallucinated-ci","publisher":"Digest AI","title":"Researchers present CiteAnchor framework for detecting hallucinated citations","datePublished":"2026-10-23T04:00:00Z","url":"https://digestai.news/story/researchers-present-citeanchor-framework-for-detecting-hallucinated-ci"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}