{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-09-30-012",
    "source_story_id": "tmp-story-fracgen-physics-video",
    "edition_id": "mp-2026-09-30-morning-0083",
    "edition_url": "https://themachinepress.com/edition/2026-09-30",
    "position": 12,
    "story_type": "dispatch",
    "section": "media-creative-tools",
    "editorial_classification": "editorial",
    "headline": "A Video Model Learned Where Materials Would Tear",
    "slug": "a-video-model-learned-where-materials-would-tear",
    "dek": "FracGen predicted RGB frames and dense physical fields together, with controls for crack origin and propagation.",
    "summary": "FracGen predicted RGB frames and dense physical fields together, with controls for crack origin and propagation.",
    "body_text": "FracGen trains on FracSim, which adds a continuum-damage model to Material Point Method simulation and produces paired fracture videos with pixel-aligned physical fields. The generator predicts those fields alongside RGB video and uses physics-informed losses to keep them consistent. The authors report material-specific stretching and tearing, plus controls for tear location, crack speed and pre-failure deformation, and say it beats tested video baselines in physical and visual fidelity. It remains a learned approximation evaluated on the paper's benchmark, not an engineering fracture solver.",
    "why_it_matters": "FracGen predicted RGB frames and dense physical fields together, with controls for crack origin and propagation.",
    "limitations": [
      "It remains a learned approximation evaluated on the paper's benchmark, not an engineering fracture solver."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-30-012/a-video-model-learned-where-materials-would-tear",
    "json_url": "https://themachinepress.com/story/mp-2026-09-30-012.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-2026-09-30-012-001",
        "text": "FracGen predicted RGB frames and dense physical fields together, with controls for crack origin and propagation.",
        "source_ids": [
          "source-2026-09-30-012"
        ],
        "qualification": "It remains a learned approximation evaluated on the paper's benchmark, not an engineering fracture solver."
      }
    ],
    "source_ids": [
      "source-2026-09-30-012"
    ],
    "tags": [
      "video generation",
      "fracture simulation",
      "physics-informed models"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-09-30-012",
      "title": "arXiv preprint 2609.38152",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2609.38152",
      "canonical_url": "https://arxiv.org/abs/2609.38152",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-09-29T13:58:31.000-04:00",
      "accessed_at": "2026-09-30T08:24:00.000-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-09-30-012-001"
      ]
    }
  ],
  "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 Video Model Learned Where Materials Would Tear",
    "publisher": "The Machine Press",
    "published_at": "2026-09-30T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-30-012/a-video-model-learned-where-materials-would-tear"
  }
}