{"ID":2922199,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T21:14:49.797962054Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.00898","arxiv_id":"2606.00898","title":"Citation Grounding: Detecting and Reducing LLM Citation Hallucinations via Legal Citation Graphs","abstract":"Large language models systematically hallucinate legal citations -- fabricating statute references, citing repealed provisions, and confusing jurisdictions -- yet no automated method exists to measure or reduce this behavior at scale. We propose citation grounding (CG), a metric that verifies LLM-generated legal citations against a ground-truth citation graph extracted from 100.8 million Ukrainian court decisions (502 million edges, 21,736 unique statute nodes). CG decomposes into three components -- citation precision (does the cited provision exist?), citation relevance (is it contextually appropriate?), and citation temporality (was it valid at the relevant date?) -- enabling differential diagnosis of hallucination types. Empirical evaluation on 100 Ukrainian legal queries across five systems -- four commercial LLMs via AWS Bedrock (Claude Haiku 4.5, Mistral Pixtral Large, Amazon Nova Pro/Lite) and one RAG-augmented production system -- reveals CG ranging from 0.791 to 0.873, with 13-21% of citations hallucinated. To reduce hallucinations without human annotation, we introduce Citation Grounding DPO (CG-DPO): a method that constructs preference pairs algorithmically by corrupting verified citations from real court decisions via four targeted strategies. On a dataset of 2,244 court decisions, a Qwen2.5-7B-Instruct model fine-tuned with LoRA achieves 98.5% mean validation accuracy in distinguishing correct from corrupted citations (rewards margin +14.9, std \u003c 0.3 pp across 3 seeds). The citation graph, evaluation framework, and CG-DPO dataset are released as open resources.","short_abstract":"Large language models systematically hallucinate legal citations -- fabricating statute references, citing repealed provisions, and confusing jurisdictions -- yet no automated method exists to measure or reduce this behavior at scale. We propose citation grounding (CG), a metric that verifies LLM-generated legal citati...","url_abs":"https://arxiv.org/abs/2606.00898","url_pdf":"https://arxiv.org/pdf/2606.00898v1","authors":"[\"Volodymyr Ovcharov\"]","published":"2026-05-30T21:22:47Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.DL\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false}