{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "bardes-et-al-vjepa-feature-prediction-2024",
    "title": "Bardes et al.: V-JEPA video feature prediction",
    "summary": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks.",
    "boundary": "Unsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.",
    "area": "Vision, robotics, reasoning and inference systems",
    "topic": "Artificial intelligence",
    "year": 2024,
    "lane": "HUMAN",
    "result": "EMPIRICAL",
    "package_id": "bardes-et-al-vjepa-feature-prediction-2024",
    "target": {
      "record_id": "urn:uuid:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
      "version": 1,
      "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
    },
    "humans": [
      "Adrien Bardes",
      "Quentin Garrido",
      "Jean Ponce",
      "Xinlei Chen",
      "Michael Rabbat",
      "Yann LeCun",
      "Mahmoud Assran",
      "Nicolas Ballas"
    ],
    "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-2404-08471v1",
        "title": "Revisiting Feature Prediction for Learning Visual Representations from Video",
        "url": "https://arxiv.org/abs/2404.08471v1"
      },
      {
        "id": "cur20261015-support-bardes-et-al-vjepa-feature-prediction-2024",
        "title": "Bardes et al.: V-JEPA video feature prediction — official research record",
        "url": "https://github.com/facebookresearch/jepa"
      }
    ],
    "points": "105000",
    "published_grade": {
      "id": "bardes-et-al-vjepa-feature-prediction-2024",
      "entry_id": "bardes-et-al-vjepa-feature-prediction-2024",
      "band": "A",
      "modifier": "-",
      "target": {
        "record_id": "urn:uuid:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
        "version": 1,
        "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
      },
      "what_changed": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks.",
      "scope": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks. Unsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.",
      "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:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
        "version": 1,
        "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
      },
      "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": "Adrien Bardes",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Quentin Garrido",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jean Ponce",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xinlei Chen",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Michael Rabbat",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yann LeCun",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Mahmoud Assran",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Nicolas Ballas",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
            "url": "https://arxiv.org/abs/2404.08471v1"
          }
        },
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-1",
          "kind": "source_reference",
          "source": {
            "citation": "Bardes et al.: V-JEPA video feature prediction — official research record",
            "url": "https://github.com/facebookresearch/jepa"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Unsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.",
        "domain": "Vision, robotics, reasoning and inference systems"
      },
      "sources": [
        {
          "citation": "Revisiting Feature Prediction for Learning Visual Representations from Video",
          "url": "https://arxiv.org/abs/2404.08471v1"
        },
        {
          "citation": "Bardes et al.: V-JEPA video feature prediction — official research record",
          "url": "https://github.com/facebookresearch/jepa"
        }
      ],
      "statement": "# Bardes et al.: V-JEPA video feature prediction\n\nLearned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks.\n\n## Scientific scope\n\nUnsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.\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\nAdrien Bardes; Quentin Garrido; Jean Ponce; Xinlei Chen; Michael Rabbat; Yann LeCun; Mahmoud Assran; Nicolas Ballas\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- [Revisiting Feature Prediction for Learning Visual Representations from Video](https://arxiv.org/abs/2404.08471v1)\n- [Bardes et al.: V-JEPA video feature prediction — official research record](https://github.com/facebookresearch/jepa)\n",
      "summary": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks.",
      "title": "Bardes et al.: V-JEPA video feature prediction"
    },
    "record_id": "urn:uuid:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
      "version": 1,
      "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:14f4f0e47634422e9e62407ace01542b57a50782177f59a04ddcdbf3617304c4"
  },
  "target": {
    "record_id": "urn:uuid:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
    "version": 1,
    "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "bardes-et-al-vjepa-feature-prediction-2024",
    "entry_id": "bardes-et-al-vjepa-feature-prediction-2024",
    "revision": 1,
    "title": "Bardes et al.: V-JEPA video feature prediction",
    "context": "Vision, robotics, reasoning and inference systems",
    "track": "Human-designed research method or resource",
    "band": "A",
    "modifier": "-",
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks.",
    "scope": "Learned visual representations from video through feature prediction without pixel reconstruction, text labels, negative examples or pretrained image encoders, evaluating frozen backbones on appearance and motion tasks. Unsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.",
    "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. Unsupervised pretraining does not remove downstream supervised evaluation or all inductive biases. This is representation learning, not the later V-JEPA2 action-conditioned robot planner.",
    "references": [
      {
        "title": "Revisiting Feature Prediction for Learning Visual Representations from Video",
        "url": "https://arxiv.org/abs/2404.08471v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      },
      {
        "title": "Bardes et al.: V-JEPA video feature prediction — official research record",
        "url": "https://github.com/facebookresearch/jepa",
        "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:dc020d87-46f8-4a53-aeda-b49ae443e5bc",
      "version": 1,
      "content_hash": "sha256:a7494632329a36353690913413560173755cf2409cbbe53bf81660fd28ceb6b6"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "105000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}