让临床文本分类模型自动决定何时用大模型，提升准确率同时控制成本
L2D-Clinical: Learning to Defer for Adaptive Model Selection in Clinical Text Classification
- 根据不确定性和文本特征，智能判断何时让小模型转交大模型处理
- 在药物不良反应检测上提纯1.7点，治疗结果预测提纯9.3点
- 适合需要高精度且关注成本的医疗AI场景
临床文本分类需在专用微调模型（如BERT）与通用大语言模型（LLM）间抉择，但二者无一在所有情况下占优。本文提出L2D-Clinical框架，学习在不确定性信号和文本特征驱动下，何时由BERT模型主动移交任务给LLM。不同于以往将决策权交给始终更优的人类专家，本方法实现自适应移交，在LLM能补足BERT短板时提升整体准确率。在两个英文临床任务上评估：(1) ADE检测（ADE Corpus V2），BioBERT F1=0.911优于LLM的F1=0.765；(2) 治疗结果分类（基于MIMIC-IV多大模型共识标注），GPT-5-nano F1=0.967优于ClinicalBERT F1=0.887。在ADE任务中，L2D-Clinical取得F1=0.928（+1.7点），仅7%样本移交，利用了LLM高召回优势；在MIMIC任务中，达到F1=0.980（+9.3点），仅16.8%样本移交。核心洞察是：模型学会精准调用大模型长处，同时降低API使用成本。
原文摘要 · Abstract (English)
Clinical text classification requires choosing between specialized fine-tuned models (BERT variants) and general-purpose large language models (LLMs), yet neither dominates across all instances. We introduce Learning to Defer for clinical text (L2D-Clinical), a framework that learns when a BERT classifier should defer to an LLM based on uncertainty signals and text characteristics. Unlike prior L2D work that defers to human experts assumed universally superior, our approach enables adaptive deferral-improving accuracy when the LLM complements BERT. We evaluate on two English clinical tasks: (1) ADE detection (ADE Corpus V2), where BioBERT (F1=0.911) outperforms the LLM (F1=0.765), and (2) treatment outcome classification (MIMIC-IV with multi-LLM consensus ground truth), where GPT-5-nano (F1=0.967) outperforms ClinicalBERT (F1=0.887). On ADE, L2D-Clinical achieves F1=0.928 (+1.7 points over BERT) by selectively deferring 7% of instances where the LLM's high recall compensates for BERT's misses. On MIMIC, L2D-Clinical achieves F1=0.980 (+9.3 points over BERT) by deferring only 16.8\% of cases to the LLM. The key insight is that L2D-Clinical learns to selectively leverage LLM strengths while minimizing API costs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。