{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "black-et-al-pi0-flow-action-expert-2024", "title": "Black et al.: pretrained vision-language control with a flow-action expert", "summary": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation.", "boundary": "Vision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.", "area": "Generalist vision-language-action systems", "topic": "Artificial intelligence", "year": 2024, "lane": "HUMAN", "result": "EMPIRICAL", "package_id": "black-et-al-pi0-flow-action-expert-2024", "target": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "humans": [ "Kevin Black", "Noah Brown", "Danny Driess", "Adnan Esmail", "Michael Equi", "Chelsea Finn", "Niccolo Fusai", "Lachy Groom", "Karol Hausman", "Brian Ichter", "Szymon Jakubczak", "Tim Jones", "Liyiming Ke", "Sergey Levine", "Adrian Li-Bell", "Mohith Mothukuri", "Suraj Nair", "Karl Pertsch", "Lucy Xiaoyang Shi", "James Tanner", "Quan Vuong", "Anna Walling", "Haohuan Wang", "Ury Zhilinsky" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261015-arxiv-2410-24164v1", "title": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } ], "points": "130000", "published_grade": { "id": "black-et-al-pi0-flow-action-expert-2024", "entry_id": "black-et-al-pi0-flow-action-expert-2024", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "what_changed": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation.", "scope": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation. Vision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Kevin Black", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Noah Brown", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Danny Driess", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Adnan Esmail", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Michael Equi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Chelsea Finn", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Niccolo Fusai", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Lachy Groom", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Karol Hausman", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Brian Ichter", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Szymon Jakubczak", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Tim Jones", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Liyiming Ke", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sergey Levine", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Adrian Li-Bell", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Mohith Mothukuri", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Suraj Nair", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Karl Pertsch", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Lucy Xiaoyang Shi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "James Tanner", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Quan Vuong", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Anna Walling", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Haohuan Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ury Zhilinsky", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "organization", "name": "Physical Intelligence", "note": "Cited historical collaboration, not a platform organization account; individual attribution remains incomplete.", "roles": [ "discoverer" ], "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "Vision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.", "domain": "Generalist vision-language-action systems" }, "sources": [ { "citation": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1" } ], "statement": "# Black et al.: pretrained vision-language control with a flow-action expert\n\nCoupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation.\n\n## Scientific scope\n\nVision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nKevin Black; Noah Brown; Danny Driess; Adnan Esmail; Michael Equi; Chelsea Finn; Niccolo Fusai; Lachy Groom; Karol Hausman; Brian Ichter; Szymon Jakubczak; Tim Jones; Liyiming Ke; Sergey Levine; Adrian Li-Bell; Mohith Mothukuri; Suraj Nair; Karl Pertsch; Lucy Xiaoyang Shi; James Tanner; Quan Vuong; Anna Walling; Haohuan Wang; Ury Zhilinsky\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [$π_0$: A Vision-Language-Action Flow Model for General Robot Control](https://arxiv.org/abs/2410.24164v1)\n", "summary": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation.", "title": "Black et al.: pretrained vision-language control with a flow-action expert" }, "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "source_ref": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:2846a46a2ef493e3dfca36d84f0701d7a45987c2e6e457cbd16fada62d5e1d2b" }, "target": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "black-et-al-pi0-flow-action-expert-2024", "entry_id": "black-et-al-pi0-flow-action-expert-2024", "revision": 1, "title": "Black et al.: pretrained vision-language control with a flow-action expert", "context": "Generalist vision-language-action systems", "track": "Human-designed research method or resource", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation.", "scope": "Coupled a pretrained vision-language backbone to a separate continuous-action flow-matching expert and trained across dexterous robot embodiments, supporting direct prompting and task-specific adaptation. Vision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Vision-language pretraining, flow matching and multi-robot policies are inherited. System integration is acquired, not invention of these components. Selected laundry/box tasks do not establish arbitrary-task autonomy, physical safety or out-of-distribution guarantees.", "references": [ { "title": "$π_0$: A Vision-Language-Action Flow Model for General Robot Control", "url": "https://arxiv.org/abs/2410.24164v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:26356e0a-955b-4665-ac31-2dfe1f54ffab", "version": 1, "content_hash": "sha256:614ee293285ac95b5429c3949cdadeb5c7a0074aa297637ecb4b35e88635e601" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "130000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }