[论文] DeepSciVerify: Verifying Scientific Claim--Citation Alignment via...

论文概要

研究领域: NLP 作者: Shaghayegh Sadeghi, Khashayar Khajavi, Rise Adhikari, et al. 发布时间: 2026-05-28 arXiv: 2605.27710

中文摘要

主张与引用证据之间的不对齐是大语言模型生成报告中的常见失败模式,限制了其在科学和其他高风险场景中的可靠性。本文提出了DeepSciVerify--一个科学主张-引用验证的两阶段管道,结合摘要级推理与选择性升级至段落级证据。系统首先利用摘要验证主张,对不确定案例推迟判断,仅在必要时检索和分析全文段落。该设计利用了不同LLM之间的互补行为:某些模型更保守,而另一些在不确定性下更果断。在SCitance基准上,DeepSciVerify达到86.7的Micro-F1,超越强摘要基线+4.5分,同时67%的实例无需全文检索即可解决。结果表明选择性证据升级同时提升了主张-引用验证的准确性和效率。

原文摘要

Misalignment between claims and their cited evidence is a common failure mode in reports generated by large language models, limiting their reliability in scientific and other high-stakes settings. We present DeepSciVerify, a two-stage pipeline for scientific claim-citation verification that combines abstract-level reasoning with selective escalation to passage-level evidence. The system first verifies claims using the abstract and defers uncertain cases, retrieving and analyzing full-text passages only when necessary. This design leverages complementary behaviors across LLMs, as some models are more conservative while others are more decisive under uncertainty. On the SCitance benchmark, DeepSciVerify achieves 86.7 Micro-F1, outperforming strong abstract-only baselines by +4.5 points while resolving 67% of实例 without full-text retrieval. These results suggest that selective evidence esca...

--- *自动采集于 2026-05-29*

#论文 #arXiv #NLP #小凯