arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.09185cs.CL
收藏

DEO:无训练直接嵌入优化用于否定感知检索

DEO: Training-Free Direct Embedding Optimization for Negation-Aware Retrieval

Department of Industrial Engineering, Hanyang University(工业工程系,翰阳大学);Department of Applied Data Science, Sungkyunkwan University(应用数据科学系,成均馆大学)

Taegyeong Lee, Jiwon Park, Seunghyun Hwang, JooYoung Jang

AI总结:

DEO通过无训练的直接嵌入优化方法,提升了否定和排除感知检索的性能,尤其在多模态检索中表现突出。

摘要:

近年来,大型语言模型(LLMs)和检索增强生成(RAG)的进步使多种检索方法成为可能。然而,现有检索方法在处理否定和排除查询时往往无法准确检索结果。为解决这一限制,先前的方法依赖于嵌入适应或微调,这会引入额外的计算成本和部署复杂性。我们提出了直接嵌入优化(DEO),一种无训练的否定感知文本和多模态检索方法。DEO将查询分解为正向和负向组成部分,并通过对比目标优化查询嵌入。在不额外使用训练数据或模型更新的情况下,DEO在NegConstraint上优于基线,获得了+0.0738 nDCG@10和+0.1028 MAP@100的提升,同时在多模态检索中将Recall@5提高了6%超过OpenAI CLIP。这些结果证明了DEO在现实应用场景中对否定和排除感知检索的实用性。

英文摘要:

Recent advances in Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) have enabled diverse retrieval methods. However, existing retrieval methods often fail to accurately retrieve results for negation and exclusion queries. To address this limitation, prior approaches rely on embedding adaptation or fine-tuning, which introduce additional computational cost and deployment complexity. We propose Direct Embedding Optimization (DEO), a training-free method for negation-aware text and multimodal retrieval. DEO decomposes queries into positive and negative components and optimizes the query embedding with a contrastive objective. Without additional training data or model updates, DEO outperforms baselines on NegConstraint, with gains of +0.0738 nDCG@10 and +0.1028 MAP@100, while improving Recall@5 by +6\% over OpenAI CLIP in multimodal retrieval. These results demonstrate the practicality of DEO for negation- and exclusion-aware retrieval in real-world settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑