C-Pack: Packed Resources For General Chinese Embeddings

By Shitao Xiao, Zheng Liu, Peitian Zhang, Niklas Muennighoff, Defu Lian, Jian-yun Nie. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2023

720 citations recorded by the source corpus.

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

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/95dda1571f1a33245726fb53e9d5060ac7a0af25

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yanzhao Zhang
2. Xin Zhang
3. Meishan Zhang
4. Dingkun Long
5. Hong-Liu Cao

Citations in the corpus

• mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval (2024, methodology): “6 epoch (250k steps) on sampled data (by XLM sampling Eq.1).”
• Mixture-of-Experts Language Models Can Be Strong and Efficient Retrievers (2026, unknown): “Dense retrievers instead encode queries and documents into vector representations and rank them by similarity (Karpukhin et al., 2020), and have become a strong approach to first-stage retrieval (Izacard et al., 2022; Wang et al., 2022; Xiao et al., 2024).”
• SkMTEB: Slovak Massive Text Embedding Benchmark and Model Adaptation (2026, unknown): “While MTEB primarily targets English, it catalyzed the development of language-specific benchmarks providing comparable depth: C-MTEB (Xiao et al., 2023) for Chinese (35 datasets), PL-MTEB (Po´swiata et al., 2024) for Polish (28 datasets), the Scandinavian Embed-ding Benchmark (Enevoldsen et al.,…”
• IdioLink: Retrieving Meaning Beyond Words Across Idiomatic and Literal Expressions (2026, unknown): “BGE models (Xiao et al., 2024) and their multilingual version M3 (Dense) (Chen et al., 2024) also fall into this category, although they are trained with only one instruction for each broad task (retrieval, clustering, etc.) We also evaluate decoder models.”
• On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey (2025, unknown): “Four typical applications of text embedding.”
• Chunk Twice, Embed Once: A Systematic Study of Segmentation and Representation Trade-offs in Chemistry-Aware Retrieval-Augmented Generation (2025, unknown): “Downstream generation was not evaluated in this study.”
• MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search (2026, unknown): “The four embedding families are MPNet [Reimers and Gurevych, 2019] ( D = 768 ), Mxbai [Lee et al., 2024] ( D = 1024 ), Nomic [Nussbaum et al., 2024] ( D = 768 ), and BGE [Xiao et al., 2023] ( D = 768 ).”
• Embedding Models Measure in Peculiar Ways (2026, unknown): “5 (Xiao et al., 2024) and its multilingual successor bge-m3 (Chen et al., 2024); mxbai-embed-large-v1 (Lee et al., 2024), which is trained with the AnglE objective (Li and Li, 2024); and the first release of the multilingual Granite embedding models ( granite-embedding-107m-multilingual and…”
• Characterizing LLM scientific concept generation: A multi-dimensional measurement study (2026, unknown): “The evolution from models like Sentence-BERT [19] and all-mpnet-base-v2 to modern architectures including BGE [21], Nomic Embed [22], and GTE [23] has improved representation quality but also introduced model-dependent variation in distance distributions.”
• REIGN: Refurbished Embeddings with Integrated Guidance Networks for Efficient Context-Length Scaling (2026, unknown): “5B-v5, Zhang et al., 2024), the GN backbones (GTE, Li et al., 2023b; BGE, Xiao et al., 2024) used directly under truncated and chunked mean-pool protocols, and a multi-GN REIGN sweep.”
• Giga-Embeddings: Mixture-of-Experts Encoders for High-Throughput Text Embeddings (2026, unknown): “E5 (Wang et al., 2022), GTE (Li et al., 2023), BGE (Xiao et al., 2023), and Nomic Embed (Nussbaum et al., 2024) scaled data and negative mining.”
• Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity (2026, unknown): “Standard vision encoders trained on natural images tend to collapse template-based documents into indistinguishable representations, failing to resolve fine-grained semantic differences that exist only in textual content [12,78].”
• When Is 0.1% Enough? Analyzing the Combined Effects of Dimensionality Reduction and Quantization on Text Embedding Compression (2026, unknown): “…models have been developed using large-scale contrastive learning, weak supervision, task-specific prefixes, or natural language instructions (Wang et al., 2022; Su et al., 2023; Asai et al., 2023; Li et al., 2023; Xiao et al., 2024; Nussbaum et al., 2025; Wang et al., 2024; Lee et al., 2025a).”
• Efficient Temporal-aware Matryoshka Adaptation for Temporal Information Retrieval (2026, unknown): “5 (bge) (Xiao et al., 2024), and bge-m3 (bgem3) (Chen et al., 2024).”
• NeurIDA: Dynamic Modeling for Effective In-Database Analytics (2025, unknown): “Implementation details could be found in https://github.com/nusdbsystem/NeurIDA.”