arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.10380cs.CLcs.AI
收藏

基于分解的声明验证中的对齐瓶颈

The Alignment Bottleneck in Decomposition-Based Claim Verification

Queen Mary University of London(伦敦女王学院);University of Bologna(博洛尼亚大学);University of Warwick(沃里克大学);The Alan Turing Institute(艾伦·图灵研究所)

Mahmud Elahi Akhter, Federico Ruggeri, Iman Munire Bilal, Rob Procter, Maria Liakata

AI总结:

本文研究了基于分解的声明验证中证据对齐和子声明错误特征的影响,发现细粒度对齐的证据能显著提升性能,而重复声明级证据则常导致性能下降。

摘要:

结构化声明分解常被提出作为验证复杂、多维度声明的解决方案,但实证结果却存在不一致。我们认为这些不一致源于两个被忽视的瓶颈:证据对齐和子声明错误特征。为了更好地理解这些因素,我们引入了一个包含真实世界复杂声明的新数据集,其特征是时间受限的证据和人工标注的子声明证据跨度。我们评估了分解在两种证据对齐设置下的表现:子声明对齐证据(SAE)和重复声明级证据(SRE)。我们的结果表明,分解只有在证据细粒度且严格对齐时才能带来显著的性能提升。相比之下,依赖重复声明级证据(SRE)的标准设置未能提升性能,甚至在不同数据集和领域(PHEMEPlus、MMM-Fact、COVID-Fact)中表现更差。此外,我们证明在存在噪声的子声明标签情况下,错误的性质最终决定了下游的鲁棒性。我们发现保守的

英文摘要:

Structured claim decomposition is often proposed as a solution for verifying complex, multi-faceted claims, yet empirical results have been inconsistent. We argue that these inconsistencies stem from two overlooked bottlenecks: evidence alignment and sub-claim error profiles. To better understand these factors, we introduce a new dataset of real-world complex claims, featuring temporally bounded evidence and human-annotated sub-claim evidence spans. We evaluate decomposition under two evidence alignment setups: Sub-claim Aligned Evidence (SAE) and Repeated Claim-level Evidence (SRE). Our results reveal that decomposition brings significant performance improvement only when evidence is granular and strictly aligned. By contrast, standard setups that rely on repeated claim-level evidence (SRE) fail to improve and often degrade performance as shown across different datasets and domains (PHEMEPlus, MMM-Fact, COVID-Fact). Furthermore, we demonstrate that in the presence of noisy sub-claim labels, the nature of the error ends up determining downstream robustness. We find that conservative "abstention" significantly reduces error propagation compared to aggressive but incorrect predictions. These findings suggest that future claim decomposition frameworks must prioritize precise evidence synthesis and calibrate the label bias of sub-claim verification models.

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。

↑