小模型交叉编码器在医疗程序重排序上胜过大模型指令微调，效率更高。
Listwise Cross-Encoder Fine-Tuning vs. Agentic Instruction Tuning for LLM Rerankers: A Systematic Study in Medical Procedure Reranking
- 用列表式学习重排序训练小规模交叉编码器，分层冻结提升效果
- 109万参数模型在NDCG@3和斯皮尔曼相关性上领先40亿参数模型2.6和13.3点
- 适合医疗信息检索场景，尤其关注高效部署与可复现性的团队
将医疗程序与患者查询进行重排序是健康保险信息检索的关键环节，因患者语言与临床术语间存在显著词汇鸿沟而复杂。本文系统比较了两种重排序范式：(1) 在多层冻结配置下，使用ListNet目标对小型交叉编码器（MedCPT、MiniLM-L12）进行微调；(2) 采用40亿参数的Qwen3-Reranker-4B指令重排序模型，其提示通过由GPT-4.1驱动的智能体优化循环迭代改进。在涵盖708项保险服务的2,647个查询构建的数据集上，109万参数的交叉编码器在NDCG@3上比40亿参数模型高2.6个百分点，在斯皮尔曼相关性上高13.3个百分点，仅需其1/37的参数量。报告了可落地的实践发现、可扩展的大型语言模型数据构建流程及部署权衡，项目代码与示例数据集已开源，支持复现与跨领域适配。
原文摘要 · Abstract (English)
Reranking medical procedures against patient queries is a critical component of health insurance information retrieval, complicated by a substantial lexical gap between patient language and clinical nomenclature. We present a systematic comparison of two reranking paradigms for this production task: (1) small cross-encoders (MedCPT, MiniLM-L12) fine-tuned with listwise learning-to-rank objectives across layer freezing configurations, and (2) Qwen3-Reranker-4B, a 4B-parameter instruction reranker whose prompt is iteratively refined via an agentic optimization loop driven by GPT-4.1. On a purpose-built dataset of 2,647 queries across 708 insurance services, we find that a 109M-parameter cross-encoder fine-tuned with ListNet outperforms the 4B-parameter model by 2.6 percentage points on NDCG@3 and 13.3 points on Spearman correlation - at 37x fewer parameters. We report practical findings, a scalable LLM based dataset construction pipeline, and deployment trade-offs relevant to production reranking systems. We release our code and a sample dataset to support reproducibility and adaptation to other domains.
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。