Metraipolis › archive › dc_422a35faa476
Unknown is not normal: separating language-model extraction from rule-based decision logic for clinical risk scores — arXiv:2609.34112
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34112
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:02:08.324Z
- SHA-256 of the archived text
565f85fc2ce408adb811964bdba3be8a2771ff5a3ca83269e737358685582611- Stable URL of this page
https://metraipolis.com/documento/dc_422a35faa476
How to cite
Unknown is not normal: separating language-model extraction from rule-based decision logic for clinical risk scores — arXiv:2609.34112 — arXiv (en), https://arxiv.org/abs/2609.34112 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_422a35faa476 (Metraipolis archive dc_422a35faa476, sha256 565f85fc2ce408adb811964bdba3be8a2771ff5a3ca83269e737358685582611)
BibTeX
@misc{dc422a35faa476,
title = {Unknown is not normal: separating language-model extraction from rule-based decision logic for clinical risk scores — arXiv:2609.34112},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34112},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_422a35faa476; sha256 565f85fc2ce408adb811964bdba3be8a2771ff5a3ca83269e737358685582611},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34112. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) are increasingly used to compute clinical risk scores from free-text notes. Notes are often incomplete, and treating undocumented findings as normal can silently misclassify patients. We test whether separating three-state extraction (present, absent or unknown, by an LLM) from decision logic (deterministic code computing score bounds over unknown inputs) lets a system ask only questions that can change the decision. On 1,200 synthetic emergency cases across six calculators (HEART, CURB-65, qSOFA, PERC, Wells, Cockcroft-Gault), with a simulated clinician answering questions, we compared this bounds policy with asking for every missing input, a missing-equals-normal schema, and an end-to-end LLM agent (Claude Opus 5.5). With Claude Haiku 4.5 as extractor, the bounds policy matched ask-all accuracy (99.4% vs 99.4%) with half the questions (0.92 vs 1.78 per case) and no irrelevant ones. Treating missing as normal dropped accuracy to 91.2% and under-triaged 8.5% of patients (95% CI 7.1-10.2), and under-triage persisted under messy notes and a noisy clinician. The agent was equally accurate under ideal conditions (99.6%) but 9.5% of its questions were irrelevant; with a noisy clinician it was less accurate than the bounds policy (83.5% vs 87.0%, p<0.001) and committed prematurely in 2.7% of cases (bounds: 0%). A 9B local model as extractor reached oracle-level accuracy (99.8%). In 584 real case reports from MedCalc-Bench, only 52% contained enough information to determine the category (HEART 13%). Routing decisions through code that reasons explicitly about unknowns avoids premature commitment and irrelevant questions, halves the questions asked, and works with small local models. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.34112 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T20:29:02.858Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).