引擎盖下：使用诊断分类器调查和改进语言模型如何跟踪一致性信息
Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Information
摘要：
神经语言模型如何跟踪主语和动词之间的数字一致性？我们表明，“诊断分类器”，训练预测的内部状态的语言模型的数量，提供了一个详细的了解如何，何时，以及在何处表示这些信息。此外，当语言模型最终出现一致性错误时，它们让我们了解数字信息何时何地被破坏。为了证明我们发现的表征所起的因果作用，我们使用一致性信息来影响LSTM在处理困难句子时的过程。这种干预的结果显示语言模型的准确性大大提高。总之，这些结果表明，诊断分类器为我们提供了神经模型中语言信息表示的无与伦比的详细研究，并证明这些知识可用于提高其性能。
英文摘要：
How do neural language models keep track of number agreement between subject and verb? We show that `diagnostic classifiers', trained to predict number from the internal states of a language model, provide a detailed understanding of how, when, and where this information is represented. Moreover, they give us insight into when and where number information is corrupted in cases where the language model ends up making agreement errors. To demonstrate the causal role played by the representations we find, we then use agreement information to influence the course of the LSTM during the processing of difficult sentences. Results from such an intervention reveal a large increase in the language model's accuracy. Together, these results show that diagnostic classifiers give us an unrivalled detailed look into the representation of linguistic information in neural models, and demonstrate that this knowledge can be used to improve their performance.
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。
