{
  "url": "https://appliedfrontier.nl/en/research/signal-dc6e4af55ac4f5d3/",
  "title": "Factorized Representations of Concurrent Temporal Variables in Pretrained Language Models",
  "published": "2026-10-29",
  "excerpt": "Language models often need to maintain multiple contextual variables that evolve independently within a document. Prior work has identified structured representations of individual temporal or",
  "content_sha256": "5e7117f82a75adb13abbb96b860232e0c1a7902e809f26e62e1dd27c7449bda5",
  "extracted_chars": 1217,
  "artifact_links": [
    "https://arxiv.org/abs/2610.40328"
  ],
  "arxiv_id": "2610.40328",
  "extraction_scope": "preprint_abstract_metadata",
  "full_text_available": false,
  "id": "signal-dc6e4af55ac4f5d3",
  "org": "Research community",
  "topic": "AI models",
  "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": [
    "AI models"
  ],
  "checked_at": "2026-11-01T12:17:04+00:00",
  "first_seen": "2026-10-31T19:32:55+00:00",
  "source_hash": "9a75a75be95e22c60b537204763d145662cb34dd0719d85f874db7ed31cb940f",
  "source_url": "https://arxiv.org/abs/2610.40328",
  "language": "en",
  "indexing": "noindex_until_editorially_enriched"
}