aiqbench analysis
This paper is primarily about applying large language models to automatic ICD coding, so AI is central rather than incidental. SemICD replaces raw ICD code strings with hierarchically clustered Semantic IDs as generation targets, plus auxiliary tasks to align those IDs with clinical meaning. The reported Micro-F1 gains of 1.34 to 8.00 points on MIMIC-III and MIMIC-IV are a useful signal, but they come from benchmark experiments, so real-world clinical value still needs validation.
Beyond ICD Codes: Semantic IDs for Generative Medical Coding
arXiv:2610.19301v1 Announce Type: new Abstract: Generative approaches to automatic International Classification of Diseases (ICD) coding use large language models (LLMs) to predict ICD codes from clinical documents. However, original ICD code strings are not ideal generation targets because the ICD hierarchy does not always reflect clinical relationships, and LLMs struggle to interpret and generate these codes. We introduce SemICD, a framework that trains LLMs to generate Semantic IDs (SIDs) instead. SemICD constructs SIDs by hierarchically clustering embeddings of textual code profiles, allowing semantically related codes to share prefixes while each complete SID uniquely identifies one ICD code. To align the SID tokens with clinical semantics, we employ two auxiliary tasks for bidirectional generation between textual code profiles and SIDs. Experiments on MIMIC-III and MIMIC-IV with four LLMs show that SemICD outperforms direct code generation, improving Micro-F1 by 1.34-8.00 points. Our code is available at https://github.com/youwillbethebest/SemICD.
Abstract:Generative approaches to automatic International Classification of Diseases (ICD) coding use large language models (LLMs) to predict ICD codes from clinical documents. However, original ICD code strings are not ideal generation targets because the ICD hierarchy does not always reflect clinical relationships, and LLMs struggle to interpret and generate these codes. We introduce SemICD, a framework that trains LLMs to generate Semantic IDs (SIDs) instead. SemICD constructs SIDs by hierarchically clustering embeddings of textual code profiles, allowing semantically related codes to share prefixes while each complete SID uniquely identifies one ICD code. To align the SID tokens with clinical semantics, we employ two auxiliary tasks for bidirectional generation between textual code profiles and SIDs. Experiments on MIMIC-III and MIMIC-IV with four LLMs show that SemICD outperforms direct code generation, improving Micro-F1 by 1.34-8.00 points. Our code is available at this https URL.
Artificial Intelligence (cs.AI)
arXiv:2610.19301 [cs.AI]
(or arXiv:2610.19301v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.19301 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 236 words
Related coverage
Beyond Visual Cues: CoT-Enhanced Reasoning for Semi-supervised Medical Image Segmentation
arXiv · Jun 17, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Can Post-Training Turn LLMs into Good Medical Coders? An Empirical Study of Generative ICD Coding
arXiv · Jun 15, 2026
Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers
Hugging Face · Aug 26, 2026