arXiv cs.AI· Wanyin Wu, Kanxue Li, Baosheng Yu, Haoyun Zhao, Yibing Zhan, Dapeng Tao, Hua Jin·· 11 小时前AI 评分20
PREBA：用 PCA 加权检索增强 LLM 与贝叶斯平均预测手术时长
PREBA: Surgical Duration Prediction via PCA-Weighted Retrieval-Augmented LLMs and Bayesian Averaging Aggregation
AI 导读
PREBA 是一个无需任务微调的检索增强框架，通过 PCA 加权检索与贝叶斯平均将 LLM 推理锚定在机构特定临床证据上。在四个真实数据集上，基于 Qwen3、DeepSeek-R1 和 HuatuoGPT-o1 的评测显示，其 MAE 最多降低 64.9%，R2 最高达 0.63，优于零样本推理并在部分数据集上超过全监督 ML 方法。
正文 · 原文
Abstract:Accurate prediction of surgical duration is pivotal for hospital resource management. Although recent supervised learning approaches, from machine learning (ML) to fine-tuned large language models (LLMs), have shown strong performance, they remain constrained by the need for high-quality labeled data and computationally intensive training. In contrast, zero-shot LLM inference offers a promising training-free alternative, but it lacks grounding in institution-specific clinical evidence, making its predictions clinically misaligned and prone to instability. To address these limitations, we present PREBA, a retrieval-augmented framework that integrates PCA-weighted retrieval and Bayesian averaging to ground LLM reasoning in institution-specific evidence without requiring task-specific fine-tuning. Specifically, PREBA first encodes heterogeneous clinical features into a unified representation. It then performs PCA-weighted retrieval, weighting each dimension by its contribution to the variation across institutional cases, instead of treating them as equally informative, and conditions the LLM to generate multiple candidate predictions. Finally, Bayesian averaging consolidates these predictions with population-level statistical priors, yielding a calibrated estimate that remains stable and clinically plausible. We evaluate PREBA on four real-world datasets spanning different institutions and languages, using three state-of-the-art LLMs, including Qwen3, DeepSeek-R1, and HuatuoGPT-o1. PREBA consistently outperforms zero-shot inference, reducing MAE by up to 64.9% and raising R2 to as high as 0.63. It further attains accuracy competitive with, and on several datasets superior to, fully supervised ML methods, while offering greater interpretability through retrieved exemplars and explicit priors. The code is available at this https URL.
| Comments: | Accepted by IEEE Journal of Biomedical and Health Informatics |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2603.13275 [cs.LG] |
| (or arXiv:2603.13275v4 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2603.13275 arXiv-issued DOI via DataCite |
Submission history
From: Kanxue Li [view email]
[v1]
Fri, 27 Feb 2026 07:19:23 UTC (1,923 KB)
[v2]
Thu, 19 Mar 2026 01:00:17 UTC (1 KB) (withdrawn)
[v3]
Sat, 21 Mar 2026 05:30:04 UTC (3,296 KB)
[v4]
Wed, 4 Nov 2026 01:22:54 UTC (2,183 KB)
来源：arXiv cs.AI · arxiv.org