{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/stochastic-attention-head-removal-a-simple","title":"Stochastic Attention Head Removal: A simple and effective method for improving Transformer Based ASR Models","arxiv_id":"2011.04004","date":"2020-11-08","proceeding":null,"authors":["Shucong Zhang","Erfan Loweimi","Peter Bell","Steve Renals"],"abstract":"Recently, Transformer based models have shown competitive automatic speech recognition (ASR) performance. One key factor in the success of these models is the multi-head attention mechanism. However, for trained models, we have previously observed that many attention matrices are close to diagonal, indicating the redundancy of the corresponding attention heads. We have also found that some architectures with reduced numbers of attention heads have better performance. Since the search for the best structure is time prohibitive, we propose to randomly remove attention heads during training and keep all attention heads at test time, thus the final model is an ensemble of models with different architectures. The proposed method also forces each head independently learn the most useful patterns. We apply the proposed method to train Transformer based and Convolution-augmented Transformer (Conformer) based ASR models. Our method gives consistent performance gains over strong baselines on the Wall Street Journal, AISHELL, Switchboard and AMI datasets. To the best of our knowledge, we have achieved state-of-the-art end-to-end Transformer based model performance on Switchboard and AMI.","url_abs":"https://arxiv.org/abs/2011.04004v2","url_pdf":"https://arxiv.org/pdf/2011.04004v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"stochastic-attention-head-removal-a-simple","repo_url":"https://github.com/s1603602/attention_head_removal","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"stochastic-attention-head-removal-a-simple","repo_url":"https://github.com/MindCode-4/code-9/tree/main/stochastic-attention-head-removal-a-simple","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"stochastic-attention-head-removal-a-simple","repo_url":"https://github.com/nanzhaogang/contrib/tree/master/application/similarity-of-neural-network-representations-revisited","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"stochastic-attention-head-removal-a-simple","repo_url":"https://github.com/pwc-1/Paper-9/tree/main/3/stochastic-attention-head-removal-a-simple","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"stochastic-attention-head-removal-a-simple","repo_url":"https://github.com/pwc-1/Paper-9/tree/main/4/stochastic-attention-head-removal-a-simple","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2011.04004","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2011.04004"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pwc-1/Paper-9/tree/main/3/stochastic-attention-head-removal-a-simple","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pwc-1/Paper-9/tree/main/4/stochastic-attention-head-removal-a-simple","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-9/tree/main/stochastic-attention-head-removal-a-simple","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/s1603602/attention_head_removal","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nanzhaogang/contrib/tree/master/application/similarity-of-neural-network-representations-revisited","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"7bd44936d5a0eb81","entry":"remove_head","repo":"s1603602/attention_head_removal","repo_kind":"official","path":"head_removal.py","file_url":"https://github.com/s1603602/attention_head_removal/blob/HEAD/head_removal.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7bd44936d5a0eb81"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}