用贝叶斯粒子滤波优化信息抽取的上下文学习，提升模型稳定性与泛化性。
BCL: Bayesian In-Context Learning Framework for Information Extraction
- 通过贝叶斯更新与粒子滤波迭代优化标签表示
- 在序列标注与关系分类任务上均实现显著且一致的性能提升
- 适合需要稳定跨规模模型表现的信息抽取场景
现有信息抽取任务越来越多地采用大语言模型的上下文学习（ICL）。然而，当前方法在不同模型规模下表现不一，缺乏系统性优化与泛化能力。为此，我们提出BCL（贝叶斯上下文学习框架），首个利用粒子滤波与贝叶斯更新系统优化信息抽取任务中标签表示的框架。通过初始化、观测、权重更新和重采样四步流程，BCL可泛化至序列标注与关系分类两种范式。大量实验表明，其在多个任务上均显著优于现有方法，且性能稳定一致。
原文摘要 · Abstract (English)
Existing information extraction (IE) tasks increasingly adopt in-context learning (ICL) with large language models. However, current approaches either show inconsistent performance across model scales or lack systematic optimization and generalizability. Building on this, we propose BCL (Bayesian In-Context Learning Framework for Information Extraction), the first optimization framework that uses particle filtering with Bayesian updates to systematically refine label representations across IE tasks. Through four steps initialization, observation, weight update, and resampling, BCL generalizes to both sequence labeling and relation classification paradigms. Extensive experiments demonstrate substantial and consistent improvements over existing approaches.
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。