{"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/vatt-transformers-for-multimodal-self","title":"VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text","arxiv_id":"2104.11178","date":"2021-04-22","proceeding":"NeurIPS 2021 12","authors":["Hassan Akbari","Liangzhe Yuan","Rui Qian","Wei-Hong Chuang","Shih-Fu Chang","Yin Cui","Boqing Gong"],"abstract":"We present a framework for learning multimodal representations from unlabeled data using convolution-free Transformer architectures. Specifically, our Video-Audio-Text Transformer (VATT) takes raw signals as inputs and extracts multimodal representations that are rich enough to benefit a variety of downstream tasks. We train VATT end-to-end from scratch using multimodal contrastive losses and evaluate its performance by the downstream tasks of video action recognition, audio event classification, image classification, and text-to-video retrieval. Furthermore, we study a modality-agnostic, single-backbone Transformer by sharing weights among the three modalities. We show that the convolution-free VATT outperforms state-of-the-art ConvNet-based architectures in the downstream tasks. Especially, VATT's vision Transformer achieves the top-1 accuracy of 82.1% on Kinetics-400, 83.6% on Kinetics-600, 72.7% on Kinetics-700, and 41.1% on Moments in Time, new records while avoiding supervised pre-training. Transferring to image classification leads to 78.7% top-1 accuracy on ImageNet compared to 64.7% by training the same Transformer from scratch, showing the generalizability of our model despite the domain gap between videos and images. VATT's audio Transformer also sets a new record on waveform-based audio event recognition by achieving the mAP of 39.4% on AudioSet without any supervised pre-training. VATT's source code is publicly available.","url_abs":"https://arxiv.org/abs/2104.11178v3","url_pdf":"https://arxiv.org/pdf/2104.11178v3.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":"vatt-transformers-for-multimodal-self","repo_url":"https://github.com/google-research/google-research/tree/master/vatt","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":null},{"paper_slug":"vatt-transformers-for-multimodal-self","repo_url":"https://github.com/akashe/ProgrammingInterview","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"vatt-transformers-for-multimodal-self","repo_url":"https://github.com/MindCode-4/code-13/tree/main/vat","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"vatt-transformers-for-multimodal-self","repo_url":"https://github.com/MindCode-4/code-9/tree/main/vat","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"vatt-transformers-for-multimodal-self","repo_url":"https://github.com/pwc-1/Paper-9/tree/main/3/vat","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"action-recognition-in-videos-2","task_name":"Action Recognition In Videos"},{"task_slug":"audio-classification","task_name":"Audio Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"},{"task_slug":"text-to-video-retrieval","task_name":"Text to Video Retrieval"},{"task_slug":"video-retrieval","task_name":"Video Retrieval"},{"task_slug":"zero-shot-video-retrieval","task_name":"Zero-Shot Video Retrieval"},{"task_slug":"image-classification","task_name":"image-classification"}],"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"},{"method_slug":"vatt","method_name":"VATT"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[{"slug":"vatt","name":"VATT","full_name":"VATT"}],"results":[{"leaderboard":"/sota/action-classification-on-kinetics-400","task":"Action Classification","dataset":"Kinetics-400","model":"VATT-Large","rank_in_archive_order":80,"of":207,"metrics":{"Acc@1":"82.1","Acc@5":"95.5"},"uses_additional_data":false},{"leaderboard":"/sota/action-classification-on-kinetics-600","task":"Action Classification","dataset":"Kinetics-600","model":"VATT-Large","rank_in_archive_order":37,"of":65,"metrics":{"Top-1 Accuracy":"83.6","Top-5 Accuracy":"96.6"},"uses_additional_data":true},{"leaderboard":"/sota/action-classification-on-moments-in-time","task":"Action Classification","dataset":"MiT","model":"VATT-Large","rank_in_archive_order":8,"of":29,"metrics":{"Top 1 Accuracy":"41.1","Top 5 Accuracy":"67.7"},"uses_additional_data":true},{"leaderboard":"/sota/audio-classification-on-audioset","task":"Audio Classification","dataset":"AudioSet","model":"VATT-Base","rank_in_archive_order":46,"of":51,"metrics":{"AUC":"0.971","Test mAP":"0.394","d-prime":"2.895"},"uses_additional_data":true},{"leaderboard":"/sota/zero-shot-video-retrieval-on-msr-vtt","task":"Zero-Shot Video Retrieval","dataset":"MSR-VTT","model":"VATT-MBS","rank_in_archive_order":41,"of":41,"metrics":{"text-to-video Median Rank":"49","text-to-video R@10":"29.7"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-retrieval-on-youcook2","task":"Zero-Shot Video Retrieval","dataset":"YouCook2","model":"VATT-MBS","rank_in_archive_order":9,"of":9,"metrics":{"text-to-video Mean Rank":"13","text-to-video R@10":"45.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2104.11178","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2104.11178"}},"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/google-research/google-research/tree/master/vatt","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-13/tree/main/vat","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-9/tree/main/vat","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/akashe/ProgrammingInterview","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pwc-1/Paper-9/tree/main/3/vat","reach":null}],"summary":{"ran":5,"unverified":3},"by_repo_kind":{"listed":{"samples":8,"ran":5,"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":8,"samples":[{"code_sha256_prefix":"36dddf850392c9f1","entry":"CrossAttentionLayer","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"36dddf850392c9f1"}},{"code_sha256_prefix":"8c095e2ffdb707ce","entry":"EncodingLayer","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8c095e2ffdb707ce"}},{"code_sha256_prefix":"4fc719fecddeb1b5","entry":"FeedForwardComponent","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4fc719fecddeb1b5"}},{"code_sha256_prefix":"d28a636f0e0774cf","entry":"MultiHeadedAttentionComponent","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d28a636f0e0774cf"}},{"code_sha256_prefix":"f9e503a955e5ba83","entry":"PositionalEncodingComponent","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f9e503a955e5ba83"}},{"code_sha256_prefix":"ef862f801aad89b6","entry":"AudioRepresentations","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.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":"ef862f801aad89b6"}},{"code_sha256_prefix":"a4a4a79006b894f0","entry":"Model","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.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":"a4a4a79006b894f0"}},{"code_sha256_prefix":"f24e8abe6f1b5c57","entry":"TextRepresentations","repo":"akashe/ProgrammingInterview","repo_kind":"listed","path":"model.py","file_url":"https://github.com/akashe/ProgrammingInterview/blob/HEAD/model.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":"f24e8abe6f1b5c57"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}