{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
    "title": "Physical Intelligence: heterogeneous semantic/action co-training for robot transfer",
    "summary": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments.",
    "boundary": "Uses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.",
    "area": "Vision-language-action transfer systems",
    "topic": "Artificial intelligence",
    "year": 2025,
    "lane": "HUMAN",
    "result": "EMPIRICAL",
    "package_id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
    "target": {
      "record_id": "urn:uuid:775e7903-aa33-4d63-aec2-e33cfaa1d265",
      "version": 1,
      "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
    },
    "humans": [
      "Kevin Black",
      "Noah Brown",
      "James Darpinian",
      "Karan Dhabalia",
      "Danny Driess",
      "Adnan Esmail",
      "Michael Equi",
      "Chelsea Finn",
      "Niccolo Fusai",
      "Manuel Y. Galliker",
      "Dibya Ghosh",
      "Lachy Groom",
      "Karol Hausman",
      "Brian Ichter",
      "Szymon Jakubczak",
      "Tim Jones",
      "Liyiming Ke",
      "Devin LeBlanc",
      "Sergey Levine",
      "Adrian Li-Bell",
      "Mohith Mothukuri",
      "Suraj Nair",
      "Karl Pertsch",
      "Allen Z. Ren",
      "Lucy Xiaoyang Shi",
      "Laura Smith",
      "Jost Tobias Springenberg",
      "Kyle Stachowicz",
      "James Tanner",
      "Quan Vuong",
      "Homer Walke",
      "Anna Walling",
      "Haohuan Wang",
      "Lili Yu",
      "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-2504-16054v1",
        "title": "$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization",
        "url": "https://arxiv.org/abs/2504.16054v1"
      }
    ],
    "points": "125000",
    "published_grade": {
      "id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
      "entry_id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:775e7903-aa33-4d63-aec2-e33cfaa1d265",
        "version": 1,
        "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
      },
      "what_changed": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments.",
      "scope": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments. Uses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.",
      "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:775e7903-aa33-4d63-aec2-e33cfaa1d265",
        "version": 1,
        "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
      },
      "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": "organization",
          "name": "Physical Intelligence",
          "note": "Cited collaboration; the full byline is retained in the structured atlas and imported ledger, committed by hash in the narrative.",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization",
            "url": "https://arxiv.org/abs/2504.16054v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization",
            "url": "https://arxiv.org/abs/2504.16054v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Uses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.",
        "domain": "Vision-language-action transfer systems"
      },
      "sources": [
        {
          "citation": "$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization",
          "url": "https://arxiv.org/abs/2504.16054v1"
        }
      ],
      "statement": "# Physical Intelligence: heterogeneous semantic/action co-training for robot transfer\n\nCo-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments.\n\n## Scientific scope\n\nUses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.\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; James Darpinian; Karan Dhabalia; Danny Driess; Adnan Esmail; Michael Equi; Chelsea Finn; Niccolo Fusai; Manuel Y. Galliker; Dibya Ghosh; Lachy Groom; Karol Hausman; Brian Ichter; Szymon Jakubczak; Tim Jones; Liyiming Ke; Devin LeBlanc; Sergey Levine; Adrian Li-Bell; Mohith Mothukuri; Suraj Nair; Karl Pertsch; Allen Z. Ren; Lucy Xiaoyang Shi; Laura Smith; Jost Tobias Springenberg; Kyle Stachowicz; James Tanner; Quan Vuong; Homer Walke; Anna Walling; Haohuan Wang; Lili Yu; 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- [$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization](https://arxiv.org/abs/2504.16054v1)\n",
      "summary": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments.",
      "title": "Physical Intelligence: heterogeneous semantic/action co-training for robot transfer"
    },
    "record_id": "urn:uuid:775e7903-aa33-4d63-aec2-e33cfaa1d265",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:775e7903-aa33-4d63-aec2-e33cfaa1d265",
      "version": 1,
      "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:2c13e995e498e032d026fa104c1c5d636d056d9351b80079dafc57d8c46cdb92"
  },
  "target": {
    "record_id": "urn:uuid:775e7903-aa33-4d63-aec2-e33cfaa1d265",
    "version": 1,
    "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
    "entry_id": "physical-intelligence-heterogeneous-vla-cotraining-2025",
    "revision": 1,
    "title": "Physical Intelligence: heterogeneous semantic/action co-training for robot transfer",
    "context": "Vision-language-action transfer 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": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments.",
    "scope": "Co-trained on robot trajectories, web vision-language tasks, object grounding and semantic subtask prediction, combining high-level instructions and a low-level action expert to transfer manipulation to unseen household environments. Uses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.",
    "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. Uses prior pi0 architecture. Acquired delta is heterogeneous-task/hybrid-example integration and transfer evidence, not another size or invention of VLA. Evaluated new homes are a finite test set, not all open worlds; later knowledge-insulation method has separate original credit.",
    "references": [
      {
        "title": "$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization",
        "url": "https://arxiv.org/abs/2504.16054v1",
        "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:775e7903-aa33-4d63-aec2-e33cfaa1d265",
      "version": 1,
      "content_hash": "sha256:c0691a5e0f1295cab5f1373444127cf8375a14ce36fb6253337fa24a61467094"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "125000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}