标题： jina-embeddings-v4：用于多模态多语言检索的通用嵌入 显示英文标题

标题： jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval

摘要： 我们引入了jina-embeddings-v4，这是一个具有38亿参数的多模态嵌入模型，通过一种新颖的架构统一了文本和图像表示，支持后期交互风格的单向量和多向量嵌入。 该模型结合了特定任务的低秩适应（LoRA）适配器，以优化在各种检索场景中的性能，包括查询-文档检索、语义文本相似性和代码搜索。 全面评估表明，jina-embeddings-v4在单模态和跨模态检索任务中均达到最先进水平，尤其在处理视觉丰富内容方面表现出色，如表格、图表、示意图和混合媒体格式。 为了便于评估此能力，我们还引入了Jina-VDR，一个专门为视觉丰富图像检索设计的新基准。 显示英文摘要

摘要： We introduce jina-embeddings-v4, a 3.8 billion parameter multimodal embedding model that unifies text and image representations through a novel architecture supporting both single-vector and multi-vector embeddings in the late interaction style. The model incorporates task-specific Low-Rank Adaptation (LoRA) adapters to optimize performance across diverse retrieval scenarios, including query-document retrieval, semantic text similarity, and code search. Comprehensive evaluations demonstrate that jina-embeddings-v4 achieves state-of-the-art performance on both single-modal and cross-modal retrieval tasks, with particular strength in processing visually rich content such as tables, charts, diagrams, and mixed-media formats. To facilitate evaluation of this capability, we also introduce Jina-VDR, a novel benchmark specifically designed for visually rich image retrieval.