{
  "url": "https://appliedfrontier.nl/en/research/signal-c8da1995d1125b97/",
  "title": "Difficulty Calibration Across Time for Membership Inference in Diffusion Models",
  "published": "2026-10-29",
  "excerpt": "Modern diffusion models are trained on massive datasets that often include copyrighted or private images. Membership inference attacks (MIAs) address this by testing whether a",
  "content_sha256": "c7b9437af04917abd3c6a030c19ca24342dc05f87f4c5cd22d7e09f9fbf264de",
  "extracted_chars": 1625,
  "artifact_links": [
    "https://arxiv.org/abs/2610.40616"
  ],
  "arxiv_id": "2610.40616",
  "extraction_scope": "preprint_abstract_metadata",
  "full_text_available": false,
  "id": "signal-c8da1995d1125b97",
  "org": "Research community",
  "topic": "Compute & efficiency",
  "source_kind": "preprint",
  "status": "unreviewed-source-report",
  "analysis_method": "deterministic_research_triage_v2",
  "novelty_verified": false,
  "proof_checked": false,
  "peer_review_verified": false,
  "matched_terms": [
    "Compute & efficiency"
  ],
  "checked_at": "2026-11-01T13:17:05+00:00",
  "first_seen": "2026-10-31T19:32:54+00:00",
  "source_hash": "c9fdc564d8e3796ec1c69516aad2c7556d9364be2a9f4d5acbd384e4335c42e9",
  "source_url": "https://arxiv.org/abs/2610.40616",
  "language": "en",
  "indexing": "noindex_until_editorially_enriched"
}