通过可学习模块识别并重采样假负例，提升密集检索效果
RRRA: Resampling and Reranking through a Retriever Adapter
- 引入适配器动态判断硬负例是否为假负例，实现查询定制化判断
- 在标准基准上超越强基线模型，显著提升检索性能
- 适合需要精准负例筛选的检索系统研发者
在密集检索中，有效训练依赖于选择高质量的硬负例，同时避免误判。现有方法基于正例文档得分设计启发式策略识别硬负例，虽提升性能与可解释性，但此类全局、不依赖样本的策略常遗漏实例相关的假负例。为此，我们提出一种可学习的适配器模块，通过监控双编码器表示来估计硬负例实为假负例的概率。该概率动态且上下文相关建模，支持细粒度、查询特定的判断。预测得分用于两个下游组件：(1) 重采样，训练中重新加权负例；(2) 重排序，在推理时对前k个召回文档重新排序。在标准基准上的实验结果表明，该适配器增强框架持续优于强基线双编码器模型，验证了显式建模假负例在密集检索中的有效性。
原文摘要 · Abstract (English)
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.
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。