{"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/t2m-gpt-generating-human-motion-from-textual","title":"T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations","arxiv_id":"2301.06052","date":"2023-01-15","proceeding":null,"authors":["Jianrong Zhang","Yangsong Zhang","Xiaodong Cun","Shaoli Huang","Yong Zhang","Hongwei Zhao","Hongtao Lu","Xi Shen"],"abstract":"In this work, we investigate a simple and must-known conditional generative framework based on Vector Quantised-Variational AutoEncoder (VQ-VAE) and Generative Pre-trained Transformer (GPT) for human motion generation from textural descriptions. We show that a simple CNN-based VQ-VAE with commonly used training recipes (EMA and Code Reset) allows us to obtain high-quality discrete representations. For GPT, we incorporate a simple corruption strategy during the training to alleviate training-testing discrepancy. Despite its simplicity, our T2M-GPT shows better performance than competitive approaches, including recent diffusion-based approaches. For example, on HumanML3D, which is currently the largest dataset, we achieve comparable performance on the consistency between text and generated motion (R-Precision), but with FID 0.116 largely outperforming MotionDiffuse of 0.630. Additionally, we conduct analyses on HumanML3D and observe that the dataset size is a limitation of our approach. Our work suggests that VQ-VAE still remains a competitive approach for human motion generation.","url_abs":"https://arxiv.org/abs/2301.06052v4","url_pdf":"https://arxiv.org/pdf/2301.06052v4.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":"t2m-gpt-generating-human-motion-from-textual","repo_url":"https://github.com/Mael-zys/T2M-GPT","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"motion-generation","task_name":"Motion Generation"},{"task_slug":"motion-synthesis","task_name":"Motion Synthesis"}],"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":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt","method_name":"GPT"},{"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":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"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":"vq-vae","method_name":"VQ-VAE"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/motion-synthesis-on-humanml3d","task":"Motion Synthesis","dataset":"HumanML3D","model":"T2M-GPT (τ = 0.5)","rank_in_archive_order":21,"of":37,"metrics":{"Diversity":"9.761","FID":"0.116","Multimodality":"1.856","R Precision Top3":"0.775"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-humanml3d","task":"Motion Synthesis","dataset":"HumanML3D","model":"T2M-GPT (τ = 0)","rank_in_archive_order":22,"of":37,"metrics":{"Diversity":"9.844","FID":"0.140","Multimodality":"3.285","R Precision Top3":"0.685"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-humanml3d","task":"Motion Synthesis","dataset":"HumanML3D","model":"T2M-GPT (τ ∈ U[0, 1])","rank_in_archive_order":23,"of":37,"metrics":{"Diversity":"9.722","FID":"0.141","Multimodality":"1.831","R Precision Top3":"0.775"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-kit-motion-language","task":"Motion Synthesis","dataset":"KIT Motion-Language","model":"T2M-GPT  (τ ∈ U[0, 1])","rank_in_archive_order":21,"of":31,"metrics":{"Diversity":"10.921","FID":"0.514","Multimodality":"1.570","R Precision Top3":"0.745"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-kit-motion-language","task":"Motion Synthesis","dataset":"KIT Motion-Language","model":"T2M-GPT (τ = 0.5)","rank_in_archive_order":23,"of":31,"metrics":{"Diversity":"10.862","FID":"0.717","Multimodality":"1.912","R Precision Top3":"0.737"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-kit-motion-language","task":"Motion Synthesis","dataset":"KIT Motion-Language","model":"T2M-GPT (τ = 0)","rank_in_archive_order":24,"of":31,"metrics":{"Diversity":"11.198","FID":"0.737","Multimodality":"2.309","R Precision Top3":"0.716"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-motion-x","task":"Motion Synthesis","dataset":"Motion-X","model":"T2M-GPT","rank_in_archive_order":2,"of":4,"metrics":{"Diversity":"10.753","FID":"1.366","MModality":"2.356","TMR-Matching Score":"0.881","TMR-R-Precision Top3":"0.655"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2301.06052","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2301.06052"}},"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/Mael-zys/T2M-GPT","reach":null}],"summary":{"ran":4,"ran_honours":1,"unverified":2},"by_repo_kind":{"official":{"samples":7,"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":0,"samples":[{"code_sha256_prefix":"c850a045304b395b","entry":"Block","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c850a045304b395b"}},{"code_sha256_prefix":"b2522d3d54eff496","entry":"CausalCrossConditionalSelfAttention","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b2522d3d54eff496"}},{"code_sha256_prefix":"5bb175d11cfeb528","entry":"CrossCondTransHead","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"5bb175d11cfeb528"}},{"code_sha256_prefix":"b5884fc2a88a2dca","entry":"PE1d_sincos","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b5884fc2a88a2dca"}},{"code_sha256_prefix":"887cbe398f9b597f","entry":"PositionEmbedding","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"887cbe398f9b597f"}},{"code_sha256_prefix":"78f9934de35620f9","entry":"CrossCondTransBase","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"78f9934de35620f9"}},{"code_sha256_prefix":"947345f4802a5dd3","entry":"Text2Motion_Transformer","repo":"Mael-zys/T2M-GPT","repo_kind":"official","path":"models/t2m_trans.py","file_url":"https://github.com/Mael-zys/T2M-GPT/blob/HEAD/models/t2m_trans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"947345f4802a5dd3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}