Med-CoReasoner: Reducing Language Disparities in Medical Reasoning via Language-Informed Co-Reasoning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913162145562624 |
|---|---|
| author | Gao, Fan Tong, Sherry T. Sohn, Jiwoong Huang, Jiahao Jiang, Junfeng Xia, Ding Ittichaiwong, Piyalitt Veerakanjana, Kanyakorn Kim, Hyunjae Chen, Qingyu Taylor, Edison Marrese Kobayashi, Kazuma Aizawa, Akiko Li, Irene |
| author_facet | Gao, Fan Tong, Sherry T. Sohn, Jiwoong Huang, Jiahao Jiang, Junfeng Xia, Ding Ittichaiwong, Piyalitt Veerakanjana, Kanyakorn Kim, Hyunjae Chen, Qingyu Taylor, Edison Marrese Kobayashi, Kazuma Aizawa, Akiko Li, Irene |
| contents | While reasoning-enhanced large language models perform strongly on English medical tasks, a persistent multilingual gap remains, with substantially weaker reasoning in local languages, limiting equitable global medical deployment. To bridge this gap, we introduce Med-CoReasoner, a language-informed co-reasoning framework that elicits parallel English and local-language reasoning, abstracts them into structured concepts, and integrates local clinical knowledge into an English logical scaffold via concept-level alignment and retrieval. This design combines the structural robustness of English reasoning with the practice-grounded expertise encoded in local languages. To evaluate multilingual medical reasoning beyond multiple-choice settings, we construct MultiMed-X, a benchmark covering seven languages with expert-annotated long-form question answering and natural language inference tasks, comprising 350 instances per language. Experiments across three benchmarks show that Med-CoReasoner improves multilingual reasoning performance by an average of 5%, with particularly substantial gains in low-resource languages. Moreover, model distillation and expert evaluation analysis further confirm that Med-CoReasoner produces clinically sound and culturally grounded reasoning traces. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_08267 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Med-CoReasoner: Reducing Language Disparities in Medical Reasoning via Language-Informed Co-Reasoning Gao, Fan Tong, Sherry T. Sohn, Jiwoong Huang, Jiahao Jiang, Junfeng Xia, Ding Ittichaiwong, Piyalitt Veerakanjana, Kanyakorn Kim, Hyunjae Chen, Qingyu Taylor, Edison Marrese Kobayashi, Kazuma Aizawa, Akiko Li, Irene Computation and Language While reasoning-enhanced large language models perform strongly on English medical tasks, a persistent multilingual gap remains, with substantially weaker reasoning in local languages, limiting equitable global medical deployment. To bridge this gap, we introduce Med-CoReasoner, a language-informed co-reasoning framework that elicits parallel English and local-language reasoning, abstracts them into structured concepts, and integrates local clinical knowledge into an English logical scaffold via concept-level alignment and retrieval. This design combines the structural robustness of English reasoning with the practice-grounded expertise encoded in local languages. To evaluate multilingual medical reasoning beyond multiple-choice settings, we construct MultiMed-X, a benchmark covering seven languages with expert-annotated long-form question answering and natural language inference tasks, comprising 350 instances per language. Experiments across three benchmarks show that Med-CoReasoner improves multilingual reasoning performance by an average of 5%, with particularly substantial gains in low-resource languages. Moreover, model distillation and expert evaluation analysis further confirm that Med-CoReasoner produces clinically sound and culturally grounded reasoning traces. |
| title | Med-CoReasoner: Reducing Language Disparities in Medical Reasoning via Language-Informed Co-Reasoning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.08267 |