arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.07473cs.CL
收藏

定位可归因文本生成中的事实不一致性

Localizing Factual Inconsistencies in Attributable Text Generation

Bar-Ilan University(巴伊兰大学);UNC Chapel Hill(北卡罗来纳大学教堂山分校);Google Research(谷歌研究)

Arie Cattan, Paul Roit, Shiyue Zhang, David Wan, Roee Aharoni, Idan Szpektor, Mohit Bansal, Ido Dagan

AI总结:

本文提出QASemConsistency,一种基于问答对分解的细粒度形式化方法,用于定位可归因文本生成中的事实不一致性,并通过众包基准和自动检测方法验证其有效性。

摘要:

近年来,无论是在人工还是自动方式下,以不同粒度级别检测模型生成文本中的幻觉现象已引起越来越多的关注。然而,大多数现有方法无法精确地指出错误所在。在这项工作中,我们引入了QASemConsistency,一种新的形式化方法,用于在细粒度级别上定位可归因文本生成中的事实不一致性。受新戴维森形式语义学的启发,我们提出将生成的文本分解为最小的谓词-论元级别的命题,以简单的问答(QA)对形式表达,并评估每个单独的QA对是否得到可信参考文本的支持。由于每个QA对对应于谓词与论元之间的单一语义关系,QASemConsistency能够有效地定位不支持的信息。我们首先通过收集众包的细粒度一致性错误标注,证明了QASemConsistency方法在人工标注中的有效性,同时实现了显著的标注者间一致性。该基准包含超过3000个实例,涵盖可归因文本生成的多种任务。我们还表明,QASemConsistency产生的事实一致性得分与人类判断高度相关。最后,我们实现了多种自动检测局部事实不一致性的方法,包括有监督的蕴含模型和大型语言模型(LLMs)。

英文摘要:

There has been an increasing interest in detecting hallucinations in model-generated texts, both manually and automatically, at varying levels of granularity. However, most existing methods fail to precisely pinpoint the errors. In this work, we introduce QASemConsistency, a new formalism for localizing factual inconsistencies in attributable text generation, at a fine-grained level. Drawing inspiration from Neo-Davidsonian formal semantics, we propose decomposing the generated text into minimal predicate-argument level propositions, expressed as simple question-answer (QA) pairs, and assess whether each individual QA pair is supported by a trusted reference text. As each QA pair corresponds to a single semantic relation between a predicate and an argument, QASemConsistency effectively localizes the unsupported information. We first demonstrate the effectiveness of the QASemConsistency methodology for human annotation, by collecting crowdsourced annotations of granular consistency errors, while achieving a substantial inter-annotator agreement. This benchmark includes more than 3K instances spanning various tasks of attributable text generation. We also show that QASemConsistency yields factual consistency scores that correlate well with human judgments. Finally, we implement several methods for automatically detecting localized factual inconsistencies, with both supervised entailment models and LLMs.

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。

↑