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 2108.11357cs.CL
收藏

ProoFVer:用于事实验证的自然逻辑定理证明

ProoFVer: Natural Logic Theorem Proving for Fact Verification

University of Cambridge(剑桥大学);Meta AI;University College London(伦敦大学学院)

Amrith Krishna, Sebastian Riedel, Andreas Vlachos

摘要:

事实验证系统通常依赖于神经网络分类器进行准确性预测,但缺乏可解释性。本文提出了ProoFer,它使用seq2seq模型来生成基于自然逻辑的推理作为证明。这些证明由索赔和检索到的证据中的跨度之间的词汇突变组成,每个跨度都用自然逻辑运算符标记。声明的准确性完全取决于这些运算符的顺序。因此,这些证明是忠实的解释,这使得ProoFer忠实的建设。目前,ProoFer在FEVER排行榜上拥有最高的标签准确性和第二好的分数。此外,它比具有反事实实例的数据集上的下一个最佳模型提高了13.21%,证明了其鲁棒性。作为解释,证明与基于注意力的亮点相比,更好地与人类的基本原理重叠,并且证明比直接使用证据更经常地帮助人类正确地预测模型决策。

英文摘要:

Fact verification systems typically rely on neural network classifiers for veracity prediction which lack explainability. This paper proposes ProoFVer, which uses a seq2seq model to generate natural logic-based inferences as proofs. These proofs consist of lexical mutations between spans in the claim and the evidence retrieved, each marked with a natural logic operator. Claim veracity is determined solely based on the sequence of these operators. Hence, these proofs are faithful explanations, and this makes ProoFVer faithful by construction. Currently, ProoFVer has the highest label accuracy and the second-best Score in the FEVER leaderboard. Furthermore, it improves by 13.21% points over the next best model on a dataset with counterfactual instances, demonstrating its robustness. As explanations, the proofs show better overlap with human rationales than attention-based highlights and the proofs help humans predict model decisions correctly more often than using the evidence directly.

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。

↑