证据差距：基于主张条件重新评分的事实核查
The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checking
AI总结：
提出SIFT方法，通过主张条件重新评分提取的证据片段，结合WSP自动NLI检查，解决LLM事实核查中证据与主张不匹配问题，在多个基准上提升准确率。
摘要：
基于LLM的事实核查系统在标准基准上实现了高判决准确率，但经常输出支持标签，其引用的证据并不支持主张。结构化分解是检查这些证据差距的自然方式，但严格的提取协议剥离了方面所需的完整主张上下文。我们引入SIFT——对提取的证据片段进行主张条件重新评分，与完整主张对比——配合WSP（支持比例），一种自动NLI检查，验证引用的证据是否蕴含主张。我们在FEVER、SciFact、5PILS和DP上使用四个开源骨干模型进行评估。SIFT在简单分解导致高达27.6个点准确率损失的单元上恢复了准确率，同时将WSP提高到直接提示之上；WSP本身以AUC 0.92和精确度0.98与人类黄金证据校准。
英文摘要：
Fact-checking systems built on LLMs achieve high verdict accuracy on standard benchmarks, yet routinely output Supports labels whose cited evidence does not license the claim. Structured decomposition is the natural way to inspect those warrants, but rigid extraction protocols strip the full-claim context that facets need. We introduce SIFT -- claim-conditioned re-scoring of extracted evidence spans against the full claim -- paired with WSP (Warranted Supports Proportion), an automatic NLI check that the cited warrant entails the claim. We evaluate on FEVER, SciFact, 5PILS, and DP across four open-source backbones. SIFT recovers accuracy on cells where naive decomposition costs up to 27.6 points, while raising WSP above direct prompting; WSP itself calibrates against human gold evidence at AUC 0.92 and precision 0.98.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
