发表机构
National Yang Ming Chiao Tung University（国立阳明交通大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对LLMs辅助同行评审时易生成无依据幻觉的问题，提出HalluPeer基准，构建对齐三元组并开展实验，发现现有检测器难以区分幻觉与合法批评，验证了真实评审中存在相关幻觉模式。
AI 中文摘要
学术同行评审规模的不断扩大，促使人们将大型语言模型（LLMs）用作评审助手，但LLMs可能会生成流畅却无依据的主张，损害评审的可靠性。现有的幻觉基准并非为同行评审设计，同行评审中验证主张需要以冗长的技术论文为依据。我们推出HalluPeer，一个用于检测科学同行评审中幻觉的基准，提供论文内容、人工撰写的评审以及注入幻觉的评审的对齐三元组，标注了检测、分类和定位信息。我们的流程生成了针对同行评审的幻觉分类体系，识别评审上下文，并通过自动过滤注入幻觉。对12000篇论文和38000篇评审的实验表明，现有检测器难以区分幻觉与合法批评，而对真实评审的评估显示，HalluPeer定义的幻觉模式存在于真实的同行评审中，凸显了对基于来源的验证的迫切需求。
英文摘要
The growing scale of academic peer review has motivated the use of Large Language Models (LLMs) as review assistants, yet LLMs can generate fluent but unsupported claims that undermine review reliability. Existing hallucination benchmarks are not designed for peer review, where verification requires grounding claims in long, technical papers. We introduce HalluPeer, a benchmark for detecting hallucinations in scientific peer reviews, providing aligned triples of paper content, human-written reviews, and hallucination-injected reviews, annotated for detection, classification, and localization. Our pipeline induces a peer-review-specific hallucination taxonomy, identifies review contexts, and injects hallucinations with automated filtering. Experiments on 12K papers and 38K reviews show that existing detectors struggle to separate hallucinations from legitimate critique, while evaluation on authentic reviews demonstrates that HalluPeer-defined hallucination patterns occur in real peer reviews, highlighting the critical need for source-aware verification. Our project page can be found in https://github.com/Lin-TzuLing/HalluPeer.git
CommentsAccepted to EMNLP Findings 2026
