Metraipolis › archive › dc_c0ab3aaa2cbd

ClinTraceBench: Source-Verifiable Longitudinal Clinical Reasoning over EHR-Derived Dialogues — arXiv:2609.01111

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.01111
License (as archived): CC0-1.0
Archived: 2026-10-17T06:41:29.435Z
SHA-256 of the archived text: f5b42c66cb6005f118b7973c52c3b7fa0f3a1069636cb4053c76a04fc1456c31
Stable URL of this page: https://metraipolis.com/documento/dc_c0ab3aaa2cbd

How to cite

ClinTraceBench: Source-Verifiable Longitudinal Clinical Reasoning over EHR-Derived Dialogues — arXiv:2609.01111 — arXiv (en), https://arxiv.org/abs/2609.01111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c0ab3aaa2cbd (Metraipolis archive dc_c0ab3aaa2cbd, sha256 f5b42c66cb6005f118b7973c52c3b7fa0f3a1069636cb4053c76a04fc1456c31)

BibTeX

@misc{dcc0ab3aaa2cbd,
  title = {ClinTraceBench: Source-Verifiable Longitudinal Clinical Reasoning over EHR-Derived Dialogues — arXiv:2609.01111},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01111},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c0ab3aaa2cbd; sha256 f5b42c66cb6005f118b7973c52c3b7fa0f3a1069636cb4053c76a04fc1456c31},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Clinical LLM assistants must reason over multi-visit patient trajectories, yet whether the compact history representations used to scale them---retrieval, structured timelines, LLM summaries, agentic memory---preserve the longitudinal signal clinical reasoning needs has not been measured. We introduce ClinTraceBench: 385 MIMIC-IV-derived verified dialogues with event-ID provenance, a nine-task taxonomy (T1--T9), and L0--L4 deterministic + L5 human-audit validation (98.92\% agreement). We evaluate eight history representation strategies---a no-context floor, \textit{last-visit-only}, \textit{full-context}, BGE-M3 \textit{dense-retrieval}, two compression schemes, and two agentic-memory systems (\textit{Mem0}, \textit{A-Mem})---across four backbones (DeepSeek-V3, GPT-4o-mini, Haiku~4.5, Sonnet~4.6) on 6{,}271 questions: 32 cells, 200{,}672 predictions. Four findings: (SP4) a controlled T3 injection probe isolates compression-induced \textit{relation} loss---with the attribution sentence present \textit{before} construction, \textit{Mem0}, \textit{A-Mem} and \textit{llm-summary} still recover only 0--5.3\% of the injected positives; (SP1) compressed strategies pay an aggregation tax on multi-visit trends and cross-patient comparisons; (SP2) the blind-to-full gap spans $+29.8$~pp (GPT-4o-mini) to $+62.7$~pp (Haiku); (SP3) abstention scales non-monotonically with context length. On the Pareto frontier Haiku dominates Sonnet under \textit{full-context} (\$25.76 vs.\ \$106.21), inverting the ``biggest backbone wins'' heuristic.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.01111 — 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 2026-10-17T06:33:38.844Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).