{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-10-17-003",
    "source_story_id": "tmp-story-value-memory-robot-policy",
    "edition_id": "mp-2026-10-17-morning-0099",
    "edition_url": "https://themachinepress.com/edition/2026-10-17",
    "position": 3,
    "story_type": "dispatch",
    "section": "robotics",
    "editorial_classification": "editorial",
    "headline": "The Robot Remembers How Close It Was",
    "slug": "the-robot-remembers-how-close-it-was",
    "dek": "Value as Memory feeds recent value predictions back into a general policy, then updates it during real-world practice.",
    "summary": "Value as Memory feeds recent value predictions back into a general policy, then updates it during real-world practice.",
    "body_text": "Partial observability can leave a robot unsure whether a long task is advancing or repeating. Value as Memory caches recent value predictions as a compact progress signal and combines offline demonstrations, human interventions and autonomous rollouts for online reinforcement learning. On the authors' deformable-object and contact-rich tasks, value caching produced a 29.4-point mean absolute success-rate gain, while online learning added 32.8 points over matched offline training. Concurrent collection and training also completed eight times as many policy iterations under the same gradient-step budget. These results cover the reported tasks and hardware, not general household autonomy.",
    "why_it_matters": "Value as Memory feeds recent value predictions back into a general policy, then updates it during real-world practice.",
    "limitations": [
      "These results cover the reported tasks and hardware, not general household autonomy."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-17-003/the-robot-remembers-how-close-it-was",
    "json_url": "https://themachinepress.com/story/mp-2026-10-17-003.json",
    "first_published_at": "2026-10-17T09:00:00.000-04:00",
    "modified_at": "2026-10-17T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-10-17-003-001",
        "text": "Value as Memory feeds recent value predictions back into a general policy, then updates it during real-world practice.",
        "source_ids": [
          "source-2026-10-17-003"
        ],
        "qualification": "These results cover the reported tasks and hardware, not general household autonomy."
      }
    ],
    "source_ids": [
      "source-2026-10-17-003"
    ],
    "tags": [
      "robotics",
      "online learning",
      "memory"
    ],
    "image_url": "https://themachinepress.com/issues/2026-10-17/value-memory-robot.webp",
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-10-17-003",
      "title": "arXiv preprint 2610.22368, original v1",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2610.22368v1",
      "canonical_url": "https://arxiv.org/abs/2610.22368v1",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-10-15T13:59:31.000-04:00",
      "accessed_at": "2026-10-17T08:26:35.645-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-10-17-003-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "The Robot Remembers How Close It Was",
    "publisher": "The Machine Press",
    "published_at": "2026-10-17T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-17-003/the-robot-remembers-how-close-it-was"
  }
}