双子星嵌入：源自双子星的可泛化嵌入表示

Gemini Embedding: Generalizable Embeddings from Gemini

March 10, 2025

作者: Jinhyuk Lee, Feiyang Chen, Sahil Dua, Daniel Cer, Madhuri Shanbhogue, Iftekhar Naim, Gustavo Hernández Ábrego, Zhe Li, Kaifeng Chen, Henrique Schechter Vera, Xiaoqi Ren, Shanfeng Zhang, Daniel Salz, Michael Boratko, Jay Han, Blair Chen, Shuo Huang, Vikram Rao, Paul Suganthan, Feng Han, Andreas Doumanoglou, Nithi Gupta, Fedor Moiseev, Cathy Yip, Aashi Jain, Simon Baumgartner, Shahrokh Shahi, Frank Palma Gomez, Sandeep Mariserla, Min Choi, Parashar Shah, Sonam Goenka, Ke Chen, Ye Xia, Koert Chen, Sai Meher Karthik Duddu, Yichang Chen, Trevor Walker, Wenlei Zhou, Rakesh Ghiya, Zach Gleicher, Karan Gill, Zhe Dong, Mojtaba Seyedhosseini, Yunhsuan Sung, Raphael Hoffmann, Tom Duerig

cs.AI

摘要

在本报告中，我们介绍了Gemini Embedding，这是一款利用Gemini强大能力的尖端嵌入模型，Gemini是谷歌目前最先进的大型语言模型。依托Gemini固有的多语言理解和代码处理能力，Gemini Embedding能够为跨越多种语言及文本模态的文本生成高度泛化的嵌入表示。由Gemini Embedding生成的表示可预先计算，并应用于多种下游任务，包括分类、相似度计算、聚类、排序及检索。在涵盖超过250种语言、包含百余项任务的大规模多语言文本嵌入基准测试（MMTEB）中，Gemini Embedding显著超越了以往的最先进模型，展现了在嵌入质量上的显著提升。我们的统一模型在MMTEB的多语言、英语及代码基准测试中均达到了业界领先水平，展现了在广泛任务选择上的强大能力，并超越了专注于特定领域的专业模型。

English

In this report, we introduce Gemini Embedding, a state-of-the-art embedding model leveraging the power of Gemini, Google's most capable large language model. Capitalizing on Gemini's inherent multilingual and code understanding capabilities, Gemini Embedding produces highly generalizable embeddings for text spanning numerous languages and textual modalities. The representations generated by Gemini Embedding can be precomputed and applied to a variety of downstream tasks including classification, similarity, clustering, ranking, and retrieval. Evaluated on the Massive Multilingual Text Embedding Benchmark (MMTEB), which includes over one hundred tasks across 250+ languages, Gemini Embedding substantially outperforms prior state-of-the-art models, demonstrating considerable improvements in embedding quality. Achieving state-of-the-art performance across MMTEB's multilingual, English, and code benchmarks, our unified model demonstrates strong capabilities across a broad selection of tasks and surpasses specialized domain-specific models.