{"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/zero-shot-whole-body-humanoid-control-via","title":"Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models","arxiv_id":"2504.11054","date":"2025-04-15","proceeding":null,"authors":["Andrea Tirinzoni","Ahmed Touati","Jesse Farebrother","Mateusz Guzek","Anssi Kanervisto","Yingchen Xu","Alessandro Lazaric","Matteo Pirotta"],"abstract":"Unsupervised reinforcement learning (RL) aims at pre-training agents that can solve a wide range of downstream tasks in complex environments. Despite recent advancements, existing approaches suffer from several limitations: they may require running an RL process on each downstream task to achieve a satisfactory performance, they may need access to datasets with good coverage or well-curated task-specific samples, or they may pre-train policies with unsupervised losses that are poorly correlated with the downstream tasks of interest. In this paper, we introduce a novel algorithm regularizing unsupervised RL towards imitating trajectories from unlabeled behavior datasets. The key technical novelty of our method, called Forward-Backward Representations with Conditional-Policy Regularization, is to train forward-backward representations to embed the unlabeled trajectories to the same latent space used to represent states, rewards, and policies, and use a latent-conditional discriminator to encourage policies to ``cover'' the states in the unlabeled behavior dataset. As a result, we can learn policies that are well aligned with the behaviors in the dataset, while retaining zero-shot generalization capabilities for reward-based and imitation tasks. We demonstrate the effectiveness of this new approach in a challenging humanoid control problem: leveraging observation-only motion capture datasets, we train Meta Motivo, the first humanoid behavioral foundation model that can be prompted to solve a variety of whole-body tasks, including motion tracking, goal reaching, and reward optimization. The resulting model is capable of expressing human-like behaviors and it achieves competitive performance with task-specific methods while outperforming state-of-the-art unsupervised RL and model-based baselines.","url_abs":"https://arxiv.org/abs/2504.11054v1","url_pdf":"https://arxiv.org/pdf/2504.11054v1.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":"zero-shot-whole-body-humanoid-control-via","repo_url":"https://github.com/facebookresearch/humenv","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"zero-shot-whole-body-humanoid-control-via","repo_url":"https://github.com/facebookresearch/metamotivo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"humanoid-control","task_name":"Humanoid Control"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"unsupervised-reinforcement-learning","task_name":"Unsupervised Reinforcement Learning"},{"task_slug":"zero-shot-generalization","task_name":"Zero-shot Generalization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.11054","atlas_url":"https://app.syntology.ai/?focus=2504.11054","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.11054"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/facebookresearch/metamotivo","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/humenv","reach":null}],"summary":{"ran":10,"ran_draft_wrong":6,"unverified":15},"by_repo_kind":{"official":{"samples":31,"ran":16,"repositories":2}},"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":31,"samples":[{"code_sha256_prefix":"7072f819126bcba5","entry":"ActorArchiConfig","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"7072f819126bcba5"}},{"code_sha256_prefix":"f0076a0b9ee82626","entry":"ArchiConfig","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f0076a0b9ee82626"}},{"code_sha256_prefix":"da844f0b3ebf1f60","entry":"BackwardArchiConfig","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"da844f0b3ebf1f60"}},{"code_sha256_prefix":"a5a3e8c377e77a82","entry":"BackwardMap","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"a5a3e8c377e77a82"}},{"code_sha256_prefix":"1f089d1db6c11d54","entry":"Config","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"1f089d1db6c11d54"}},{"code_sha256_prefix":"c0563387a851f5be","entry":"DenseParallel","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"c0563387a851f5be"}},{"code_sha256_prefix":"d1e1861240004451","entry":"ForwardArchiConfig","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"d1e1861240004451"}},{"code_sha256_prefix":"cc09fe08b93015a6","entry":"Norm","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"cc09fe08b93015a6"}},{"code_sha256_prefix":"96fae2cd4fb681f8","entry":"ParallelLayerNorm","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"96fae2cd4fb681f8"}},{"code_sha256_prefix":"0c73c35f93c90cfb","entry":"TruncatedNormal","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"0c73c35f93c90cfb"}},{"code_sha256_prefix":"d572ba72d9339001","entry":"build_backward","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"d572ba72d9339001"}},{"code_sha256_prefix":"7bebc0e93d8ff2e2","entry":"config_from_dict","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"7bebc0e93d8ff2e2"}},{"code_sha256_prefix":"ee57a886c99dd0b1","entry":"layernorm","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"ee57a886c99dd0b1"}},{"code_sha256_prefix":"c092126f33402e08","entry":"linear","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"c092126f33402e08"}},{"code_sha256_prefix":"feddcd5c2a9fd9a2","entry":"load_model","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"feddcd5c2a9fd9a2"}},{"code_sha256_prefix":"c168a019337992da","entry":"make_from_name","repo":"facebookresearch/humenv","repo_kind":"official","path":"humenv/env.py","file_url":"https://github.com/facebookresearch/humenv/blob/HEAD/humenv/env.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OPAQUE_PARAMS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"c168a019337992da"}},{"code_sha256_prefix":"92468117ea8270ba","entry":"Actor","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"92468117ea8270ba"}},{"code_sha256_prefix":"01f0428c88bde5de","entry":"Block","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"01f0428c88bde5de"}},{"code_sha256_prefix":"79c4ebb09e02aaff","entry":"FBModel","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"79c4ebb09e02aaff"}},{"code_sha256_prefix":"4071a91883ae5542","entry":"ForwardMap","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"4071a91883ae5542"}},{"code_sha256_prefix":"e1091fe2e9a10d40","entry":"ResidualActor","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"e1091fe2e9a10d40"}},{"code_sha256_prefix":"5d792e7fe20ec8bd","entry":"ResidualBlock","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5d792e7fe20ec8bd"}},{"code_sha256_prefix":"6c14b5df35e6dc85","entry":"ResidualForwardMap","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"6c14b5df35e6dc85"}},{"code_sha256_prefix":"47676b887f26870f","entry":"SequetialFMap","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"47676b887f26870f"}},{"code_sha256_prefix":"00681738ef179d76","entry":"_build_batch_forward","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"00681738ef179d76"}},{"code_sha256_prefix":"300a59992d3edcb4","entry":"build_actor","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"300a59992d3edcb4"}},{"code_sha256_prefix":"8857dd4c88d11076","entry":"build_forward","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"8857dd4c88d11076"}},{"code_sha256_prefix":"999b9fb5750aca10","entry":"dict_to_config","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"999b9fb5750aca10"}},{"code_sha256_prefix":"68a1973e5ff83c48","entry":"eval_mode","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"68a1973e5ff83c48"}},{"code_sha256_prefix":"20a14ba3346110ba","entry":"residual_embedding","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"20a14ba3346110ba"}},{"code_sha256_prefix":"af0184a440f48156","entry":"simple_embedding","repo":"facebookresearch/metamotivo","repo_kind":"official","path":"metamotivo/fb/model.py","file_url":"https://github.com/facebookresearch/metamotivo/blob/HEAD/metamotivo/fb/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"af0184a440f48156"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}