{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-09-04-004",
    "source_story_id": "tmp-story-gift-action-sufficient-features",
    "edition_id": "mp-2026-09-04-morning-0057",
    "edition_url": "https://themachinepress.com/edition/2026-09-04",
    "position": 4,
    "story_type": "dispatch",
    "section": "robotics",
    "editorial_classification": "editorial",
    "headline": "The Robot's Rich Vision Still Forgot What Control Needed",
    "slug": "the-robot-s-rich-vision-still-forgot-what-control-needed",
    "dek": "GIFT supervises intermediate features for geometry, affordances and goal regions while leaving three different action formulations intact.",
    "summary": "GIFT supervises intermediate features for geometry, affordances and goal regions while leaving three different action formulations intact.",
    "body_text": "The authors call the mismatch between visually rich representations and control-useful structure the action-sufficiency gap. Their training constraints align geometry, predict instruction-relevant affordances and reconstruct goal regions inside a VLA policy and two world-action models. On zero-shot LIBERO-Plus transfer, the three variants gained 4.6, 12.6 and 5.2 points over matched counterparts; on RoboCasa the reported gains were 12.6, 9.0 and 8.4 points. Those results support the tested training principle, not a general guarantee for unseen robots.",
    "why_it_matters": "GIFT supervises intermediate features for geometry, affordances and goal regions while leaving three different action formulations intact.",
    "limitations": [
      "Those results support the tested training principle, not a general guarantee for unseen robots."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-04-004/the-robot-s-rich-vision-still-forgot-what-control-needed",
    "json_url": "https://themachinepress.com/story/mp-2026-09-04-004.json",
    "first_published_at": "2026-09-04T09:00:00.000-04:00",
    "modified_at": "2026-09-04T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-09-04-004-001",
        "text": "GIFT supervises intermediate features for geometry, affordances and goal regions while leaving three different action formulations intact.",
        "source_ids": [
          "source-2026-09-04-004"
        ],
        "qualification": "Those results support the tested training principle, not a general guarantee for unseen robots."
      }
    ],
    "source_ids": [
      "source-2026-09-04-004"
    ],
    "tags": [
      "robot manipulation",
      "VLA",
      "world-action models"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-09-04-004",
      "title": "arXiv preprint 2609.04193",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2609.04193",
      "canonical_url": "https://arxiv.org/abs/2609.04193",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-09-03T13:59:03.000-04:00",
      "accessed_at": "2026-09-04T08:24:03.294-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-09-04-004-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "The Robot's Rich Vision Still Forgot What Control Needed",
    "publisher": "The Machine Press",
    "published_at": "2026-09-04T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-04-004/the-robot-s-rich-vision-still-forgot-what-control-needed"
  }
}