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 2502.08080cs.CL
收藏

显微镜下的NLI:原子假设分解揭示了什么

NLI under the Microscope: What Atomic Hypothesis Decomposition Reveals

University of Maryland, College Park(马里兰大学帕克分校)

Neha Srikanth, Rachel Rudinger

AI总结:

本文利用原子假设分解将NLI和可废止NLI任务拆分为原子子问题,以探究模型逻辑一致性,发现LLM在此类细粒度推理上仍存在困难,并提出衡量模型推理一致性的新指标。

摘要:

将文本分解为原子命题是一个灵活的框架,允许对输入和输出文本进行更深入的检查。我们在两个自然语言推理任务(传统NLI和可废止NLI)中使用假设的原子分解来形成原子子问题,即模型在解决整体问题时必须权衡的细粒度推理。这些原子子问题作为一种工具,用于进一步理解NLI和可废止推理的结构,探究模型对不同推理的一致性和理解,并衡量基准数据集中示例的多样性。我们的结果表明,LLM在原子NLI和可废止NLI子问题上仍然难以保持逻辑一致性。最后,我们识别出可废止NLI示例的关键原子子问题,即对整体标签贡献最大的子问题,并提出一种方法来衡量模型的推理一致性,这是一种旨在捕捉模型在不同上下文中对同一事实做出始终正确或错误预测程度的指标。

英文摘要:

Decomposition of text into atomic propositions is a flexible framework allowing for the closer inspection of input and output text. We use atomic decomposition of hypotheses in two natural language reasoning tasks, traditional NLI and defeasible NLI, to form atomic sub-problems, or granular inferences that models must weigh when solving the overall problem. These atomic sub-problems serve as a tool to further understand the structure of both NLI and defeasible reasoning, probe a model's consistency and understanding of different inferences, and measure the diversity of examples in benchmark datasets. Our results indicate that LLMs still struggle with logical consistency on atomic NLI and defeasible NLI sub-problems. Lastly, we identify critical atomic sub-problems of defeasible NLI examples, or those that most contribute to the overall label, and propose a method to measure the inferential consistency of a model, a metric designed to capture the degree to which a model makes consistently correct or incorrect predictions about the same fact under different contexts.

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。

↑