GENIUS用生成方式实现跨模态通用检索，效率远超传统方法。
GENIUS: A Generative Framework for Universal Multimodal Search
- 将多模态数据转为带语义的离散标识符，支持跨任务跨模态检索
- 在M-BEIR上性能超越现有生成式方法，且数据库越大越快
- 适合需要高速检索的工业级应用，尤其适合多模态场景
生成式检索是信息检索中的新兴方法，通过生成目标数据的标识符（ID）实现高效搜索，可替代传统的基于嵌入的检索。然而现有模型多为任务专用，性能仍不及嵌入式方法。本文提出GENIUS，一个支持多种任务、多模态与多领域的通用生成式检索框架。核心创新是模态解耦语义量化，将多模态数据转化为编码了模态与语义的离散ID。同时引入查询增强策略，通过在查询与其目标间插值，提升模型对不同查询形式的适应能力。在M-BEIR基准上，GENIUS显著优于先前生成式方法。与嵌入式检索相比，GENIUS在数据库规模增大时仍保持高检索速度，且在多个基准上表现优异；结合重排序后，其性能接近嵌入式方法，同时保留高效优势。
原文摘要 · Abstract (English)
Generative retrieval is an emerging approach in information retrieval that generates identifiers (IDs) of target data based on a query, providing an efficient alternative to traditional embedding-based retrieval methods. However, existing models are task-specific and fall short of embedding-based retrieval in performance. This paper proposes GENIUS, a universal generative retrieval framework supporting diverse tasks across multiple modalities and domains. At its core, GENIUS introduces modality-decoupled semantic quantization, transforming multimodal data into discrete IDs encoding both modality and semantics. Moreover, to enhance generalization, we propose a query augmentation that interpolates between a query and its target, allowing GENIUS to adapt to varied query forms. Evaluated on the M-BEIR benchmark, it surpasses prior generative methods by a clear margin. Unlike embedding-based retrieval, GENIUS consistently maintains high retrieval speed across database size, with competitive performance across multiple benchmarks. With additional re-ranking, GENIUS often achieves results close to those of embedding-based methods while preserving efficiency.
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。