{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-10-03-003",
    "source_story_id": "tmp-story-interevolve-reward-programs",
    "edition_id": "mp-2026-10-03-morning-0086",
    "edition_url": "https://themachinepress.com/edition/2026-10-03",
    "position": 3,
    "story_type": "dispatch",
    "section": "robotics",
    "editorial_classification": "editorial",
    "headline": "A Humanoid Rewrote the Reward, Not Its Weights",
    "slug": "a-humanoid-rewrote-the-reward-not-its-weights",
    "dek": "InterEvolve revised staged reward programs at test time and verified candidates before retaining them.",
    "summary": "InterEvolve revised staged reward programs at test time and verified candidates before retaining them.",
    "body_text": "InterEvolve pairs an object-aware behavioral foundation model with reward programs whose stages, completion tests and constants can be revised from execution feedback. An LLM agent changes program structure while a numerical optimizer tunes constants; candidates are verified across parallel simulations before joining a skill library. The authors report diverse simulated loco-manipulation behaviors and autonomous execution of evolved skills on a physical Unitree G1, but the evidence remains specific to the tested controller, tasks and hardware.",
    "why_it_matters": "InterEvolve revised staged reward programs at test time and verified candidates before retaining them.",
    "limitations": [],
    "importance": 9,
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-03-003/a-humanoid-rewrote-the-reward-not-its-weights",
    "json_url": "https://themachinepress.com/story/mp-2026-10-03-003.json",
    "first_published_at": "2026-10-03T09:00:00.000-04:00",
    "modified_at": "2026-10-03T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-10-03-003-001",
        "text": "InterEvolve revised staged reward programs at test time and verified candidates before retaining them.",
        "source_ids": [
          "source-2026-10-03-003"
        ],
        "qualification": null
      }
    ],
    "source_ids": [
      "source-2026-10-03-003"
    ],
    "tags": [
      "humanoid robots",
      "test-time adaptation",
      "reward programs"
    ],
    "image_url": "https://themachinepress.com/issues/2026-10-03/interevolve-robot-arm.webp",
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-10-03-003",
      "title": "arXiv preprint 2610.02196",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2610.02196",
      "canonical_url": "https://arxiv.org/abs/2610.02196",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-10-01T13:59:40.000-04:00",
      "accessed_at": "2026-10-03T08:31:37.000-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-10-03-003-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "A Humanoid Rewrote the Reward, Not Its Weights",
    "publisher": "The Machine Press",
    "published_at": "2026-10-03T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-03-003/a-humanoid-rewrote-the-reward-not-its-weights"
  }
}