{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-09-30-006",
    "source_story_id": "tmp-story-simpleicl-robot-learning",
    "edition_id": "mp-2026-09-30-morning-0083",
    "edition_url": "https://themachinepress.com/edition/2026-09-30",
    "position": 6,
    "story_type": "dispatch",
    "section": "robotics",
    "editorial_classification": "editorial",
    "headline": "Robot Learning Got a Smaller In-Context Recipe",
    "slug": "robot-learning-got-a-smaller-in-context-recipe",
    "dek": "SimpleICL defined what a visual demonstration should transfer, then paired it with low-cost data collection.",
    "summary": "SimpleICL defined what a visual demonstration should transfer, then paired it with low-cost data collection.",
    "body_text": "SimpleICL separates the information packed into a robot's visual demonstration—action, semantics, affordances, spatial relations and goal—by defining the intended learning target explicitly. Its minimalist framework combines a visual prompt encoder with a low-cost collection protocol and, according to the authors, performs strongly in simulation and real-world tests without massive pretraining or specialized infrastructure. Experiments probe action, semantic, composition and affordance discrimination. The paper promises an open release, so reproducibility will depend on the final data and training pipeline.",
    "why_it_matters": "SimpleICL defined what a visual demonstration should transfer, then paired it with low-cost data collection.",
    "limitations": [
      "Its minimalist framework combines a visual prompt encoder with a low-cost collection protocol and, according to the authors, performs strongly in simulation and real-world tests without massive pretraining or specialized infrastructure."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-30-006/robot-learning-got-a-smaller-in-context-recipe",
    "json_url": "https://themachinepress.com/story/mp-2026-09-30-006.json",
    "first_published_at": "2026-09-30T09:00:00.000-04:00",
    "modified_at": "2026-09-30T09:02:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-09-30-006-001",
        "text": "SimpleICL defined what a visual demonstration should transfer, then paired it with low-cost data collection.",
        "source_ids": [
          "source-2026-09-30-006"
        ],
        "qualification": "Its minimalist framework combines a visual prompt encoder with a low-cost collection protocol and, according to the authors, performs strongly in simulation and real-world tests without massive pretraining or specialized infrastructure."
      }
    ],
    "source_ids": [
      "source-2026-09-30-006"
    ],
    "tags": [
      "in-context learning",
      "robot manipulation",
      "visual demonstrations"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-09-30-006",
      "title": "arXiv preprint 2609.38173",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2609.38173",
      "canonical_url": "https://arxiv.org/abs/2609.38173",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-09-29T13:59:45.000-04:00",
      "accessed_at": "2026-09-30T08:24:00.000-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-09-30-006-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "Robot Learning Got a Smaller In-Context Recipe",
    "publisher": "The Machine Press",
    "published_at": "2026-09-30T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-30-006/robot-learning-got-a-smaller-in-context-recipe"
  }
}