{"url":"/sota/robot-manipulation-generalization-on-the","task":{"name":"Robot Manipulation Generalization","url":"/task/robot-manipulation-generalization","note":null},"dataset":{"name":"The COLOSSEUM","url":"/dataset/the-colosseum"},"category":"Robots","categories":["Robots"],"category_note":null,"description":null,"description_from":null,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Average decrease average across all perturbations"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Average decrease average across all perturbations":null}},"counts":{"rows":9,"rows_with_code":6,"rows_with_paper_page":8,"rows_dated":8,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"RVT","metrics":{"Average decrease average across all perturbations":"-14.5"},"uses_additional_data":false,"paper_date":"2022-06-19","paper":"/paper/0-1-deep-neural-networks-via-block-coordinate","paper_url":"https://arxiv.org/abs/2206.09379v2","paper_title":"0/1 Deep Neural Networks via Block Coordinate Descent","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"ACT","metrics":{"Average decrease average across all perturbations":"-61.8"},"uses_additional_data":false,"paper_date":"2023-04-23","paper":"/paper/learning-fine-grained-bimanual-manipulation","paper_url":"https://arxiv.org/abs/2304.13705v1","paper_title":"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"GENIMA","metrics":{"Average decrease average across all perturbations":"-41.6"},"uses_additional_data":false,"paper_date":"2024-07-10","paper":"/paper/generative-image-as-action-models","paper_url":"https://arxiv.org/abs/2407.07875v2","paper_title":"Generative Image as Action Models","code":"https://github.com/MohitShridhar/genima","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"3D Diffuser Actor","metrics":{"Average decrease average across all perturbations":"-15.6"},"uses_additional_data":false,"paper_date":"2024-02-18","paper":"/paper/3d-diffuser-actor-policy-diffusion-with-3d","paper_url":"https://arxiv.org/abs/2402.10885","paper_title":"3D Diffuser Actor: Policy Diffusion with 3D Scene Representations","code":"https://github.com/nickgkan/3d_diffuser_actor","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":9,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"RVT2","metrics":{"Average decrease average across all perturbations":"-19.5"},"uses_additional_data":false,"paper_date":"2024-06-12","paper":"/paper/rvt-2-learning-precise-manipulation-from-few","paper_url":"https://arxiv.org/abs/2406.08545v1","paper_title":"RVT-2: Learning Precise Manipulation from Few Demonstrations","code":"https://github.com/NVlabs/RVT","n_code_links":1,"syntology":{"n_ran":15,"n_unverified":0,"n_samples":15,"n_pointer_only_licence":15}},{"rank_in_archive_order":6,"model":"PerAct","metrics":{"Average decrease average across all perturbations":"-17.3"},"uses_additional_data":false,"paper_date":"2022-09-12","paper":"/paper/perceiver-actor-a-multi-task-transformer-for","paper_url":"https://arxiv.org/abs/2209.05451v2","paper_title":"Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation","code":"https://github.com/peract/peract","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"R3M","metrics":{"Average decrease average across all perturbations":"-49.9"},"uses_additional_data":false,"paper_date":"2022-03-23","paper":"/paper/r3m-a-universal-visual-representation-for","paper_url":"https://arxiv.org/abs/2203.12601v3","paper_title":"R3M: A Universal Visual Representation for Robot Manipulation","code":"https://github.com/facebookresearch/r3m","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":8,"model":"MVP","metrics":{"Average decrease average across all perturbations":"-16.3"},"uses_additional_data":false,"paper_date":"2022-03-11","paper":"/paper/masked-visual-pre-training-for-motor-control","paper_url":"https://arxiv.org/abs/2203.06173v1","paper_title":"Masked Visual Pre-training for Motor Control","code":"https://github.com/ir413/mvp","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"SAM2Act","metrics":{"Average decrease average across all perturbations":"-4.3"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":4,"rows_with_any_sample_ran":2,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":18,"n_unverified":13,"n_samples":31,"n_pointer_only_licence":21,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":18,"n_unverified":13,"n_samples":31,"n_pointer_only_licence":21,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}