---
title: "BiRG-LoRA：面向跨基准医疗问答的临床结构化秩门控低秩适应"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2606.31432v1"
permalink: "https://feed.astralor.com/items/itm_8a028c0ac2454d3b7e566ecf"
published: "2026-06-30T09:59:05.000Z"
score: 75
channel: "ai"
tags: ["BiRG-LoRA","Qwen3","低秩适应","医疗问答","门控机制"]
---

# BiRG-LoRA：面向跨基准医疗问答的临床结构化秩门控低秩适应

## AI 点评

> 这篇论文对做医疗NLP和参数高效微调的研究者特别值得一看：它把临床先验直接融入LoRA的秩选择中，不是简单混合专家，而是让同一个适配器根据输入动态调整表达能力，思路很巧妙。对比实验设计严谨，虽然只用了单种子但统计检验到位。对想在实际医疗产品中低成本适配大模型的人来说，这个减参不减效的方案很有参考价值。

## 速览

本文提出BiRG-LoRA，一种单适配器秩门控LoRA方法，专用于医疗多项选择问答。它通过双轴门控机制，结合隐藏语义证据与临床专业先验，动态选择稀疏的top-k秩原子，并利用标量注入系数控制适配强度。在Qwen3-8B模型上，BiRG-LoRA在CMB、CMExam、MedQA和MedMCQA四个基准上平均准确率达69.31%，超越MoELoRA 0.89个百分点，且可训练参数减少28.1%。该方法证明了临床结构化秩分配可有效提升跨基准医疗问答性能。

## 关键点

- 提出BiRG-LoRA，单适配器但实现输入条件化的秩门控，通过双轴门控（隐藏语义+临床先验）选择稀疏秩子集，并控制适配强度。
- 在Qwen3-8B上，四个医疗问答基准平均准确率69.31%，超越MoELoRA 0.89个百分点，参数减少28.1%。
- 与普通LoRA r16及秩匹配的LoRA r4相比，BiRG-LoRA提升0.83个宏平均点，且对标签噪声不敏感。

## 全文

Medical multiple-choice question answering requires parameter-efficient adaptation across heterogeneous knowledge domains and reasoning operations. A medication question, a diagnostic decision, a public-health item, and a nursing-action item may require different low-rank updates, while some recall items should preserve the base model's representation with only mild adapter intervention. We propose BiRG-LoRA, a single-adapter rank-gated LoRA method for medical question answering. BiRG-LoRA keeps one LoRA module per target layer but makes its rank dimension input-conditioned: for each question, a biaxial gate combines hidden semantic evidence with specialty/profession priors, clinical-operation priors, and their interaction to select a sparse top-$k$ subset of rank atoms. A scalar injection coefficient further controls the strength of the selected adapter update. Under a matched Qwen3-8B CMB-source protocol, BiRG-LoRA achieves the highest four-benchmark macro-average accuracy among trainable PEFT baselines and matched routing controls: 69.31% averaged over CMB, CMExam, MedQA, and MedMCQA. It improves over MoELoRA by 0.89 percentage points while using 28.1% fewer trainable parameters; a paired, benchmark-stratified bootstrap over final predictions gives a 95% confidence interval of [0.42, 1.37] for this macro-average gain. Basic controls show that BiRG-LoRA also improves over vanilla LoRA r16 and active-rank-matched LoRA r4 by 0.83 macro points, and an evaluation-time weak-axis perturbation check suggests that performance is not brittle to moderate tag noise. The results support a bounded claim: clinically structured rank allocation improves cross-benchmark medical QA under a matched single-seed protocol, while training-seed variance remains future work.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_8a028c0ac2454d3b7e566ecf