发表机构
Faculty of Data Science, City University of Macau; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences; Southern University of Science and Technology（澳门城市大学数据科学学院; 中国科学院深圳先进技术研究院; 南方科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究旨在评估中文长篇法律研究报告引用可信度，引入LegalCiteTrust基准，含72个任务并从三个维度评估。通过实验发现不同系统表现各异，检索工具与基于E/F/A的修订效果不同，强调可靠法律研究生成需检索后的引用感知证据治理。
AI 中文摘要
长篇法律研究报告越来越依赖大语言模型和智能研究系统，但其可靠性不仅取决于任务回答，还取决于所引用法律权威的可信度。即使引用指向真实来源，也可能存在风险。我们引入LegalCiteTrust，一个评估中文长篇法律研究报告引用可信度的基准。它包含72个密集注释的报告级任务，并从覆盖范围、支持度和引用可信度三个维度评估报告。引用可信度通过引用级别的存在性、忠实性和适用性（E/F/A）来衡量。实验表明不同系统在任务完成、证据丰富度、引用密度和引用可靠性方面表现不同。检索工具可改善证据支持但不一定提高信任分数，基于E/F/A的修订比仅基于存在性的过滤更能提高信任度和最终分数。这些结果表明，可靠的法律研究生成需要检索后的引用感知证据治理。
英文摘要
Long-form legal research reports increasingly rely on LLMs and agentic research systems, but their reliability depends not only on answering the task, but also on whether cited legal authorities are trustworthy. A citation can be risky even when it points to a real source: the report may omit limiting conditions, misdescribe the authority, or use it to support a stronger claim than the source allows. We introduce LegalCiteTrust, a benchmark for evaluating citation trustworthiness in Chinese long-form legal research reports. It contains 72 densely annotated report-level tasks and evaluates reports along three dimensions: Coverage, Support, and Citation Trustworthiness. Citation Trustworthiness is operationalized through citation-level Existence, Fidelity, and Applicability (E/F/A). Experiments on general-purpose LLMs, deep-research systems, and legal-specific systems show that task completion, evidence richness, citation density, and citation reliability expose different system behaviors. Retrieval tools can improve evidence support without reliably improving the Trust score, while E/F/A-based revision improves Trust and Final score more clearly than existence-only filtering. These results suggest that trustworthy legal research generation requires citation-aware evidence governance after retrieval: systems must not only retrieve legal authorities, but also select, describe, and apply them reliably.
Comments8 pages, 21 pages with appendix, 26 tables, 4 figures
