How Far Does BERT Look At:Distance-based Clustering and Analysis of BERT$'$s Attention
This work provides a more systematic analysis of attention mechanisms in Transformer models like BERT, which is incremental as it builds on previous heuristic methods.
The authors tackled the problem of systematically analyzing BERT's attention patterns by clustering attention heatmaps using unsupervised methods, resulting in the identification of distinct patterns that align with prior observations and enabling explanation and calibration of attention heads.
Recent research on the multi-head attention mechanism, especially that in pre-trained models such as BERT, has shown us heuristics and clues in analyzing various aspects of the mechanism. As most of the research focus on probing tasks or hidden states, previous works have found some primitive patterns of attention head behavior by heuristic analytical methods, but a more systematic analysis specific on the attention patterns still remains primitive. In this work, we clearly cluster the attention heatmaps into significantly different patterns through unsupervised clustering on top of a set of proposed features, which corroborates with previous observations. We further study their corresponding functions through analytical study. In addition, our proposed features can be used to explain and calibrate different attention heads in Transformer models.