arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.08252cs.IRcs.CL
收藏

ReasonEmbed:用于推理密集型文档检索的增强型文本嵌入

ReasonEmbed: Enhanced Text Embeddings for Reasoning-Intensive Document Retrieval

University of Science and Technology of China(中国科学技术大学);Beijing Academy of Artificial Intelligence(北京人工智能研究院);Beijing University of Posts and Telecommunications(北京邮电大学);State Key Laboratory of Cognitive Intelligence(认知智能国家重点实验室);Hong Kong Polytechnic University(香港理工大学)

Jianlyu Chen, Junwei Lan, Chaofan Li, Defu Lian, Zheng Liu

AI总结:

本文提出ReasonEmbed,通过ReMixer合成数据、Redapter自适应学习算法和多规模backbone实现,提升推理密集型检索性能,其中ReasonEmbed-Qwen3-8B在BRIGHT基准上取得38.1的nDCG@10高分。

摘要:

本文介绍ReasonEmbed,一种为推理密集型文档检索设计的新型文本嵌入模型。我们的工作包含三个关键技术贡献:首先,我们提出ReMixer,一种新的数据合成方法,克服了先前合成数据集中的平凡性问题,能够生成82,000个高质量训练样本。其次,我们设计Redapter,一种自适应学习算法,根据每个样本的推理强度动态调整训练权重,使模型能有效捕捉查询与文档之间的复杂语义关系。第三,我们在多种不同规模的backbone上实现ReasonEmbed,所有模型在推理密集型检索任务中均表现优异。值得注意的是,我们的ReasonEmbed-Qwen3-8B模型在BRIGHT基准上取得了38.1的nDCG@10分数,显著优于现有文本嵌入模型。我们将在ReasonEmbed中完全开源所创建的资源,以推动该领域研究的进展。

英文摘要:

In this paper, we introduce ReasonEmbed, a novel text embedding model developed for reasoning-intensive document retrieval. Our work includes three key technical contributions. First, we propose ReMixer, a new data synthesis method that overcomes the triviality problem prevalent in previous synthetic datasets, enabling large-scale production of 82K high-quality training samples. Second, we design Redapter, a self-adaptive learning algorithm that dynamically adjusts training each sample's weight based on its reasoning intensity. This allows the model to effectively capture the complex semantic relationships between queries and documents. Third, we implement ReasonEmbed across multiple backbones of varying sizes, all of which achieve superior performance on reasoning-intensive retrieval tasks. Notably, our ReasonEmbed-Qwen3-8B model offers a record-high nDCG@10 score of 38.1 on the BRIGHT benchmark, which significantly outperforms existing text embedding models. We will fully open-source our created resources in ReasonEmbed to push forward the research advancement in this field.

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。

↑