JEDA：从环境对话中查询自由的临床订单检索
摘要
临床对话将显式指令（请求胸部X光）与隐式推理（咳嗽夜间加重，我们应该检查肺炎）混合在一起。许多系统依赖 LLM 重写，引入延迟、不稳定性和不透明性，阻碍实时下单。我们提出 JEDA（Joint Embedding for Direct and Ambient clinical orders），一个 domain-initialized bi-encoder，直接检索标准化订单，并在 query-free 模式下，对环境对话的短滚动窗口进行编码以触发检索。Initialized from PubMedBERT and fine-tuned with a duplicate-safe contrastive objective, JEDA 将异构意图表达对齐到共享订单概念。Training uses constrained LLM guidance to tie each signed order to complementary formulations (command only, context only, command+context, context+reasoning)，producing clearer inter-order separation, tighter query extendash order coupling, and stronger generalization. The query-free mode is noise-resilient, reducing sensitivity to disfluencies and ASR errors by conditioning on a short window rather than a single utterance. Deployed in practice, JEDA yields large gains and substantially outperforms its base encoder and recent open embedders (Linq Embed Mistral, SFR Embedding, GTE Qwen, BGE large, Embedding Gemma)。结果是一个 fast、可解释、LLM-free retrieval layer that links ambient context to actionable clinical orders in real time。
关键词
引用
@article{arxiv.2510.14169,
title = {JEDA: Query-Free Clinical Order Search from Ambient Dialogues},
author = {Praphul Singh and Corey Barrett and Sumana Srivasta and Amitabh Saikia and Irfan Bulu and Sri Gadde and Krishnaram Kenthapadi},
journal= {arXiv preprint arXiv:2510.14169},
year = {2025}
}