When Citations Mislead? A Claim-Level Benchmark for Legal Hallucination Detection

Abstract:Large language models are increasingly used in legal research and drafting, but they can still produce claims that sound convincing without being supported by the cited source. We introduce PARCEL, a benchmark for checking whether a legal claim is supported by the underlying authority. Using recent New York State Court of Appeals decisions, we build a dataset of 3,396 parenthetical-style claims labeled as Supported, Refuted, or Not Found. We cast this task as a three-way natural language inference problem and evaluate several state-of-the-art LLMs in a zero-shot setting. Although the strongest models reach up to 0.97 accuracy, the results also show an important weakness: models still incorrectly mark unsupported claims as supported, even when the full opinion text is provided. Across models, missing support is harder to detect than direct contradiction, and fabricated but plausible citations cause the largest drop in performance. Overall, PARCEL provides a practical benchmark for testing claim-level groundedness in legal RAG systems.

9 pages, 2 figures, 7 tables. Published in the 21st International Conference on Artificial Intelligence and Law (ICAIL 2026), Singapore. Dataset: this https URL

Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

arXiv:2610.10971 [cs.CL]

(or arXiv:2610.10971v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2610.10971 arXiv-issued DOI via DataCite (pending registration)

In Proceedings of the 21st International Conference on Artificial Intelligence and Law (ICAIL 2026), June 08-12, 2026, Singapore. ACM, New York, NY, USA

https://doi.org/10.1145/3836937.3836948 DOI(s) linking to related resources

From: M. Mikail Demir [view email] [v1] Wed, 7 Oct 2026 22:43:56 UTC (174 KB)