解析交叉编码器匹配机制，揭示注意力头的关键作用。
Understanding Matching Mechanisms in Cross-Encoders
- 通过注意力分析，定位影响匹配决策的关键神经头。
- 揭示交叉编码器内部匹配检测的内在机制。
- 适合对模型可解释性感兴趣的NLP研究者。
神经信息检索架构，尤其是交叉编码器，表现出极强的性能，但其内部机制仍不明确。现有研究多关注高层次行为（如输入内容如何影响预测、模型是否遵循已知信息检索公理），却未能深入描述匹配过程。本文不采用专门的机械可解释性方法，而是证明简单方法同样能提供重要洞察：首先，聚焦注意力过程，提取因果性见解，凸显部分注意力头在该过程中的关键作用；其次，进一步解释匹配检测背后的机制。
原文摘要 · Abstract (English)
Neural IR architectures, particularly cross-encoders, are highly effective models whose internal mechanisms are mostly unknown. Most works trying to explain their behavior focused on high-level processes (e.g., what in the input influences the prediction, does the model adhere to known IR axioms) but fall short of describing the matching process. Instead of Mechanistic Interpretability approaches which specifically aim at explaining the hidden mechanisms of neural models, we demonstrate that more straightforward methods can already provide valuable insights. In this paper, we first focus on the attention process and extract causal insights highlighting the crucial roles of some attention heads in this process. Second, we provide an interpretation of the mechanism underlying matching detection.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。