We sometimes observe monotonically decreasing cross-attention weights in our Conformer-based global attention-based encoder-decoder (AED) models, Further investigation shows that the Conformer encoder reverses the sequence in the time dimension. We analyze the initial behavior of the decoder cross-attention mechanism and find that it encourages the Conformer encoder self-attention to build a connection between the initial frames and all other informative frames. Furthermore, we show that, at some point in training, the self-attention module of the Conformer starts dominating the output over the preceding feed-forward module, which then only allows the reversed information to pass through. We propose methods and ideas of how this flipping can be avoided and investigate a novel method to obtain label-frame-position alignments by using the gradients of the label log probabilities w.r.t. the encoder input frames.
Cite
@article{arxiv.2410.00680,
title = {The Conformer Encoder May Reverse the Time Dimension},
author = {Robin Schmitt and Albert Zeyer and Mohammad Zeineldeen and Ralf Schlüter and Hermann Ney},
journal= {arXiv preprint arXiv:2410.00680},
year = {2025}
}