Step 4: tool results

get_claim_instances

{
  "claim_id": "01232bf1-55f7-42fd-9f39-8a38bfde7837"
}

{
  "claim_id": "01232bf1-55f7-42fd-9f39-8a38bfde7837",
  "total": 1,
  "limit": 20,
  "offset": 0,
  "has_more": false,
  "instances": [
    {
      "id": "787f7aab-85bf-4a52-a21d-38a98e6a7774",
      "source_id": "585b1989-2399-46f7-abbb-38babf339276",
      "verbatim_text": "Fuzzing can sometimes identify tasks where models can be made to perform better, but is unreliable and does\nnot distinguish sandbagging from benign underelicitation.",
      "context": "Section 6.5.2; noise injection into weights via LoRA adapters",
      "stance": "affirms",
      "confidence": 0.75,
      "speaker": null,
      "publication": null,
      "source_date": null,
      "link": null,
      "source_title": "Auditing Games for Sandbagging",
      "source_type": "preprint",
      "source_url": "https://arxiv.org/pdf/2512.07810",
      "source_published_date": "2025-12-08",
      "source_authors": [
        "Taylor, Jordan",
        "Black, Sid",
        "Bowen, Dillon",
        "Read, Thomas",
        "Golechha, Satvik",
        "Zelenka-Martin, Alex",
        "Makins, Oliver",
        "Kissane, Connor",
        "Ayonrinde, Kola",
        "Merizian, Jacob",
        "Marks, Samuel",
        "Cundy, Chris",
        "Bloom, Joseph"
      ],
      "source_publisher": "arXiv.org",
      "source_doi": null,
      "source_interest": null,
      "recorded_by": "extractor",
      "display_rank": null,
      "reading": null,
      "draws_on": [],
      "story": {
        "standing": "untraced",
        "basis": "none",
        "depth": 0,
        "diverges": false,
        "downstream_total": 0
      }
    }
  ]
}