cs.AI updates on arXiv.org 13小时前
RRRA: Resampling and Reranking through a Retriever Adapter
index_new5.html
../../../zaker_core/zaker_tpl_static/wap/tpl_guoji1.html

 

本文提出一种基于监测Bi-Encoder表示的学习适配器模块,用于估计密集检索中硬负样本的真实性,有效减少错误负样本,提高检索性能和可解释性。

arXiv:2508.11670v1 Announce Type: cross Abstract: In dense retrieval, effective training hinges on selecting high quality hard negatives while avoiding false negatives. Recent methods apply heuristics based on positive document scores to identify hard negatives, improving both performance and interpretability. However, these global, example agnostic strategies often miss instance specific false negatives. To address this, we propose a learnable adapter module that monitors Bi-Encoder representations to estimate the likelihood that a hard negative is actually a false negative. This probability is modeled dynamically and contextually, enabling fine-grained, query specific judgments. The predicted scores are used in two downstream components: (1) resampling, where negatives are reweighted during training, and (2) reranking, where top-k retrieved documents are reordered at inference. Empirical results on standard benchmarks show that our adapter-enhanced framework consistently outperforms strong Bi-Encoder baselines, underscoring the benefit of explicit false negative modeling in dense retrieval.

Fish AI Reader

Fish AI Reader

AI辅助创作,多种专业模板,深度分析,高质量内容生成。从观点提取到深度思考,FishAI为您提供全方位的创作支持。新版本引入自定义参数,让您的创作更加个性化和精准。

FishAI

FishAI

鱼阅,AI 时代的下一个智能信息助手,助你摆脱信息焦虑

联系邮箱 441953276@qq.com

相关标签

密集检索 Bi-Encoder 适配器模块 错误负样本 性能提升
相关文章