Metraipolis › archive › dc_ac63134fc377

Discrete Neural Interlingua for Ontology-Agnostic EHR Modeling — arXiv:2610.27047

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27047
License (as archived)
CC0-1.0
Archived
2026-10-21T07:07:27.695Z
SHA-256 of the archived text
6aa02adf05b2125acc298f4b970a95fe429919198f854e28bb6ec71bd73f48a9
Stable URL of this page
https://metraipolis.com/documento/dc_ac63134fc377

How to cite

Discrete Neural Interlingua for Ontology-Agnostic EHR Modeling — arXiv:2610.27047 — arXiv (en), https://arxiv.org/abs/2610.27047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ac63134fc377 (Metraipolis archive dc_ac63134fc377, sha256 6aa02adf05b2125acc298f4b970a95fe429919198f854e28bb6ec71bd73f48a9)
BibTeX
@misc{dcac63134fc377,
  title = {Discrete Neural Interlingua for Ontology-Agnostic EHR Modeling — arXiv:2610.27047},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27047},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ac63134fc377; sha256 6aa02adf05b2125acc298f4b970a95fe429919198f854e28bb6ec71bd73f48a9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

EHR foundation models promise scalable clinical prediction across health systems, yet their deployment is constrained by a tokenization bottleneck. The same clinical concept can appear as different ontology codes, local identifiers, synonyms, or free-text descriptions, which standard EHR tokenizers treat as unrelated symbols. A model trained on MIMIC-III with a closed-vocabulary tokenizer can use only 13% of eICU events. We introduce MEDIATOR, a discrete ontology-agnostic tokenizer that maps any clinical concept expressed as text into a shared vector-quantized codebook, turning EHR harmonization into open-vocabulary tokenization without manual mappings or a fixed vocabulary. Trained with synonym supervision and guided by enriched concept descriptions, code co-occurrence, and relation graphs, MEDIATOR collapses equivalent expressions while organizing distinct concepts in a medically meaningful discrete space. Used as the token interface of an EHR Transformer, MEDIATOR improves mean AUPRC over the strongest prior tokenizer by 4.0 points (8% relative) on 20 MIMIC-III tasks and by 8.6 points (29% relative) in zero-shot transfer to 14 eICU tasks, where it uses all events. The transfer gain combines coverage and representation, as MEDIATOR remains the best tokenizer under identical inputs (+1.0 and +1.9 points on operational and subcohort-mortality tasks). Same-encoder controls show that the text encoder drives in-domain accuracy, while the learned objectives of MEDIATOR drive transfer robustness and entity-linking quality. MEDIATOR also enables transferable attribution across hospitals.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.27047 — 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 2026-10-21T07:05:43.523Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).