Researchers present CiteAnchor framework for detecting hallucinated citations
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…
1 source primary source
Key points
- 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.
Read the original at arXiv cs.CL · by Chenyi Tong, Ping Nie, Yu Zhang primary sourceOpen source ↗
Topics · follow one to build your own front page
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- MIT historian says early AI engineers added randomness to mimic human creativity · 1 src
- Researchers introduce MemPlan for text-based planning with PDDL · 1 src
- Researchers find self‑rewarded moral argumentation makes language models decisive but pliable · 1 src
- AQL-Lite improves Arabic LLMs with curated data and partial-parameter training · 1 src
- arXiv study maps gaps in AI research for low-resource languages · 1 src
Comments
via GitHub Discussions