C-Pack: Packed Resources For General Chinese Embeddings

TL;DR

C-Pack introduces comprehensive resources including benchmarks, datasets, and models to significantly improve Chinese text embeddings, achieving state-of-the-art results and providing valuable tools for future research.

Contribution

The paper presents a new suite of resources—C-MTEB, C-MTP, and C-TEM—that advance Chinese embeddings and outperform previous models, also including English embedding resources.

Findings

C-TEM models outperform prior Chinese embeddings by up to 10%.

The English models achieve state-of-the-art performance on MTEB.

The released English data is twice as large as the Chinese data.

Abstract

We introduce C-Pack, a package of resources that significantly advance the field of general Chinese embeddings. C-Pack includes three critical resources. 1) C-MTEB is a comprehensive benchmark for Chinese text embeddings covering 6 tasks and 35 datasets. 2) C-MTP is a massive text embedding dataset curated from labeled and unlabeled Chinese corpora for training embedding models. 3) C-TEM is a family of embedding models covering multiple sizes. Our models outperform all prior Chinese text embeddings on C-MTEB by up to +10% upon the time of the release. We also integrate and optimize the entire suite of training methods for C-TEM. Along with our resources on general Chinese embedding, we release our data and models for English text embeddings. The English models achieve state-of-the-art performance on MTEB benchmark; meanwhile, our released English data is 2 times larger than the Chinese data. All these resources are made publicly available at https://github.com/FlagOpen/FlagEmbedding.