[论文解读] On the Capabilities and Limitations of Reasoning for Natural Language Understanding.

Daniel Khashabi, Erfan Sadeqi Azer+3 more|arXiv (Cornell University)|Jan 8, 2019

Natural Language Processing Techniques参考文献 32被引用 17

一句话总结

本文提出一个形式化框架，将隐藏的、无歧义的概念意义空间与嘈杂的语言符号空间联系起来，以分析自然语言理解中推理能力的局限性。该框架证明，由于噪声累积，多跳推理在超过对数级跳跃时会失败，揭示了语言模型在复杂推理方面存在的根本性障碍。

ABSTRACT

Recent systems for natural language understanding are strong at overcoming linguistic variability for lookup style reasoning. Yet, their accuracy drops dramatically as the number of reasoning steps increases. We present the first formal framework to study such empirical observations, addressing the ambiguity, redundancy, incompleteness, and inaccuracy that the use of language introduces when representing a hidden conceptual space. Our formal model uses two interrelated spaces: a conceptual meaning space that is unambiguous and complete but hidden, and a linguistic symbol space that captures a noisy grounding of the meaning space in the symbols or words of a language. We apply this framework to study the connectivity problem in undirected graphs---a core reasoning problem that forms the basis for more complex multi-hop reasoning. We show that it is indeed possible to construct a high-quality algorithm for detecting connectivity in the (latent) meaning graph, based on an observed noisy symbol graph, as long as the noise is below our quantified noise level and only a few hops are needed. On the other hand, we also prove an impossibility result: if a query requires a large number (specifically, logarithmic in the size of the meaning graph) of hops, no reasoning system operating over the symbol graph is likely to recover any useful property of the meaning graph. This highlights a fundamental barrier for a class of reasoning problems and systems, and suggests the need to limit the distance between the two spaces, rather than investing in multi-hop reasoning with many hops.

研究动机与目标

• 形式化自然语言理解系统中语言符号与潜在概念意义之间的差距。
• 分析语言表征中的噪声如何影响多步推理的准确性。
• 识别自然语言理解系统中推理的理论极限。
• 确定在何种条件下，基于符号语言的推理能够可靠地恢复隐藏概念图的属性。
• 建立多跳推理在何时变得根本不可靠的形式化边界。

提出的方法

• 引入两个相互关联的空间：隐藏的、无歧义的概念意义空间，以及代表语言的嘈杂语言符号空间。
• 将推理建模为从意义图派生出的符号图上的推断，控制噪声水平。
• 以图连通性作为核心推理问题，用于检验框架的预测。
• 根据跳跃次数量化可靠推理所允许的最大噪声水平。
• 证明一个不可能性结果：当推理需要在意义图中进行对数级跳跃时，任何系统都无法可靠地恢复意义属性。
• 将该框架应用于分析多跳推理在自然语言理解中的可行性，以图连通性作为测试案例。

实验结果

研究问题

• RQ1 在何种条件下，基于嘈杂语言符号的推理能够恢复隐藏概念意义空间的属性？
• RQ2 推理步骤数（跳跃次数）如何影响自然语言理解系统中推理的可靠性？
• RQ3 在多跳任务中，仍能保证准确推理的最大噪声水平是多少？
• RQ4 一个形式化框架能否预测多跳推理何时变得根本不可行？
• RQ5 基于语言的系统所能实现的推理深度的理论极限是什么？

主要发现

• 由于语言表征中噪声的累积，推理准确率随着步骤数的增加而显著下降。
• 当噪声低于量化阈值且仅需少量跳跃时，高质量算法能够检测意义图中的连通性。
• 当查询需要相对于意义图大小的对数级跳跃时，任何推理系统都无法可靠地恢复有用属性。
• 该不可能性结果与算法复杂度无关，表明深度多跳推理存在根本性障碍。
• 该框架识别出基于语言的系统中推理深度与噪声容忍度之间的关键权衡。
• 结果表明，投入大量跳跃次数进行多跳推理很可能无法成功，系统应限制推理距离。