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 2602.16299cs.IR
收藏

MICE:用于高效重排的最小交互交叉编码器

MICE: Minimal Interaction Cross-Encoders for efficient Re-ranking

Sinequa by ChapsVision(ChapsVision旗下Sinequa);Univ. Toulouse, IRIT, CLLE, CNRS(图卢兹大学,IRIT,CLLE,CNRS);Sorbonne Université, CNRS, ISIR(索邦大学,CNRS,ISIR)

Mathias Vast, Victor Morand, Josiane Mothe, Benjamin Piwowarski

AI总结:

研究针对交叉编码器推理成本高的问题,提出MICE架构,通过去除有害或不必要交互,将加速交叉编码器推理与提升一级检索效果两种方法结合,评估显示其降低推理延迟四倍,效果良好。

摘要:

交叉编码器在信息检索中提供了最先进的排序效果,但推理成本高。先前工作从两个不同方向解决此瓶颈。本文基于对交叉编码器内部机制的深入理解,提出将两种方法结合,得到新架构MICE,在域内和域外数据集上广泛评估,MICE推理延迟降低四倍。

英文摘要:

In Information Retrieval (IR), cross-encoders deliver state-of-the-art ranking effectiveness but have a high inference cost, limiting their use to second-stage re-rankers. Prior work has addressed this bottleneck from two largely separate directions: accelerating cross-encoder inference through attention sparsification, or improving first-stage retrieval effectiveness to alleviate the need of a re-ranker, using more complex models, e.g. late-interactions. In this work, we bridge these two directions through an in-depth analysis of cross-encoder internal mechanisms. By identifying and removing superfluous interactions, we derive MICE (Minimal Interaction Cross-Encoders), a new cross-encoder architecture that retains effectiveness while reducing computational overhead. Extensive evaluations show MICE retains most of the performances of its cross-encoder counterparts in-domain and matches or even exceeds it in out-of-domain, while reducing FLOPs down to 2.5 times.

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。

↑