Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ray, Sushant Kumar, Kashyap, Gautam Siddharth, Tripathi, Sahil, Joshi, Nipun, Govindarajan, Vijay, Ali, Rafiq, Gao, Jiechao, Naseem, Usman
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917209742245888
author Ray, Sushant Kumar
Kashyap, Gautam Siddharth
Tripathi, Sahil
Joshi, Nipun
Govindarajan, Vijay
Ali, Rafiq
Gao, Jiechao
Naseem, Usman
author_facet Ray, Sushant Kumar
Kashyap, Gautam Siddharth
Tripathi, Sahil
Joshi, Nipun
Govindarajan, Vijay
Ali, Rafiq
Gao, Jiechao
Naseem, Usman
contents Clinical Question-Answering (CQA) industry systems are increasingly rely on Large Language Models (LLMs), yet their deployment is often guided by the assumption that domain-specific fine-tuning is essential. Although specialised medical LLMs such as BioBERT, BioGPT, and PubMedBERT remain popular, they face practical limitations including narrow coverage, high retraining costs, and limited adaptability. Efforts based on Supervised Fine-Tuning (SFT) have attempted to address these assumptions but continue to reinforce what we term the SPECIALISATION FALLACY-the belief that specialised medical LLMs are inherently superior for CQA. To address this assumption, we introduce MEDASSESS-X, a deployment-industry-oriented CQA framework that applies alignment at inference time rather than through SFT. MEDASSESS-X uses lightweight steering vectors to guide model activations toward medically consistent reasoning without updating model weights or requiring domain-specific retraining. This inference-time alignment layer stabilises CQA performance across both general-purpose and specialised medical LLMs, thereby resolving the SPECIALISATION FALLACY. Empirically, MEDASSESS-X delivers consistent gains across all LLM families, improving Accuracy by up to +6%, Factual Consistency by +7%, and reducing Safety Error Rate by as much as 50%.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12812
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning?
Ray, Sushant Kumar
Kashyap, Gautam Siddharth
Tripathi, Sahil
Joshi, Nipun
Govindarajan, Vijay
Ali, Rafiq
Gao, Jiechao
Naseem, Usman
Computation and Language
Clinical Question-Answering (CQA) industry systems are increasingly rely on Large Language Models (LLMs), yet their deployment is often guided by the assumption that domain-specific fine-tuning is essential. Although specialised medical LLMs such as BioBERT, BioGPT, and PubMedBERT remain popular, they face practical limitations including narrow coverage, high retraining costs, and limited adaptability. Efforts based on Supervised Fine-Tuning (SFT) have attempted to address these assumptions but continue to reinforce what we term the SPECIALISATION FALLACY-the belief that specialised medical LLMs are inherently superior for CQA. To address this assumption, we introduce MEDASSESS-X, a deployment-industry-oriented CQA framework that applies alignment at inference time rather than through SFT. MEDASSESS-X uses lightweight steering vectors to guide model activations toward medically consistent reasoning without updating model weights or requiring domain-specific retraining. This inference-time alignment layer stabilises CQA performance across both general-purpose and specialised medical LLMs, thereby resolving the SPECIALISATION FALLACY. Empirically, MEDASSESS-X delivers consistent gains across all LLM families, improving Accuracy by up to +6%, Factual Consistency by +7%, and reducing Safety Error Rate by as much as 50%.
title Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning?
topic Computation and Language
url https://arxiv.org/abs/2601.12812