Shitao commited on
Commit
c11661b
1 Parent(s): b5c9d86

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -9
README.md CHANGED
@@ -21,18 +21,30 @@ language:
21
  <p>
22
  </h4>
23
 
24
- More details please refer to our Github: [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding).
25
 
 
26
 
27
- [English](README.md) | [中文](https://github.com/FlagOpen/FlagEmbedding/blob/master/README_zh.md)
28
 
29
- FlagEmbedding can map any text to a low-dimensional dense vector which can be used for tasks like retrieval, classification, clustering, or semantic search.
30
- And it also can be used in vector databases for LLMs.
31
 
32
- ************* 🌟**Updates**🌟 *************
33
- - 10/12/2023: Release [LLM-Embedder](./FlagEmbedding/llm_embedder/README.md), a unified embedding model to support diverse retrieval augmentation needs for LLMs. [Paper](https://arxiv.org/pdf/2310.07554.pdf) :fire:
34
- - 09/15/2023: The [technical report](https://arxiv.org/pdf/2309.07597.pdf) of BGE has been released
35
- - 09/15/2023: The [masive training data](https://data.baai.ac.cn/details/BAAI-MTP) of BGE has been released
 
 
 
 
 
 
 
 
 
 
 
 
 
36
  - 09/12/2023: New models:
37
  - **New reranker model**: release cross-encoder models `BAAI/bge-reranker-base` and `BAAI/bge-reranker-large`, which are more powerful than embedding model. We recommend to use/fine-tune them to re-rank top-k documents returned by embedding models.
38
  - **update embedding model**: release `bge-*-v1.5` embedding model to alleviate the issue of the similarity distribution, and enhance its retrieval ability without instruction.
@@ -57,6 +69,7 @@ And it also can be used in vector databases for LLMs.
57
 
58
  | Model | Language | | Description | query instruction for retrieval [1] |
59
  |:-------------------------------|:--------:| :--------:| :--------:|:--------:|
 
60
  | [BAAI/llm-embedder](https://huggingface.co/BAAI/llm-embedder) | English | [Inference](./FlagEmbedding/llm_embedder/README.md) [Fine-tune](./FlagEmbedding/llm_embedder/README.md) | a unified embedding model to support diverse retrieval augmentation needs for LLMs | See [README](./FlagEmbedding/llm_embedder/README.md) |
61
  | [BAAI/bge-reranker-large](https://huggingface.co/BAAI/bge-reranker-large) | Chinese and English | [Inference](#usage-for-reranker) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/reranker) | a cross-encoder model which is more accurate but less efficient [2] | |
62
  | [BAAI/bge-reranker-base](https://huggingface.co/BAAI/bge-reranker-base) | Chinese and English | [Inference](#usage-for-reranker) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/reranker) | a cross-encoder model which is more accurate but less efficient [2] | |
@@ -73,7 +86,6 @@ And it also can be used in vector databases for LLMs.
73
  | [BAAI/bge-base-zh](https://huggingface.co/BAAI/bge-base-zh) | Chinese | [Inference](#usage-for-embedding-model) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/finetune) | a base-scale model but with similar ability to `bge-large-zh` | `为这个句子生成表示以用于检索相关文章:` |
74
  | [BAAI/bge-small-zh](https://huggingface.co/BAAI/bge-small-zh) | Chinese | [Inference](#usage-for-embedding-model) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/finetune) | a small-scale model but with competitive performance | `为这个句子生成表示以用于检索相关文章:` |
75
 
76
-
77
  [1\]: If you need to search the relevant passages to a query, we suggest to add the instruction to the query; in other cases, no instruction is needed, just use the original query directly. In all cases, **no instruction** needs to be added to passages.
78
 
79
  [2\]: Different from embedding model, reranker uses question and document as input and directly output similarity instead of embedding. To balance the accuracy and time cost, cross-encoder is widely used to re-rank top-k documents retrieved by other simple models.
 
21
  <p>
22
  </h4>
23
 
24
+ For more details please refer to our Github: [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding).
25
 
26
+ If you are looking for a model that supports more languages, longer texts, and other retrieval methods, you can try using [bge-m3](https://huggingface.co/BAAI/bge-m3).
27
 
 
28
 
29
+ [English](README.md) | [中文](https://github.com/FlagOpen/FlagEmbedding/blob/master/README_zh.md)
 
30
 
31
+ FlagEmbedding focuses on retrieval-augmented LLMs, consisting of the following projects currently:
32
+
33
+ - **Long-Context LLM**: [Activation Beacon](https://github.com/FlagOpen/FlagEmbedding/tree/master/Long_LLM/activation_beacon)
34
+ - **Fine-tuning of LM** : [LM-Cocktail](https://github.com/FlagOpen/FlagEmbedding/tree/master/LM_Cocktail)
35
+ - **Dense Retrieval**: [BGE-M3](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/BGE_M3), [LLM Embedder](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/llm_embedder), [BGE Embedding](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/baai_general_embedding)
36
+ - **Reranker Model**: [BGE Reranker](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/reranker)
37
+ - **Benchmark**: [C-MTEB](https://github.com/FlagOpen/FlagEmbedding/tree/master/C_MTEB)
38
+
39
+ ## News
40
+ - 1/30/2024: Release **BGE-M3**, a new member to BGE model series! M3 stands for **M**ulti-linguality (100+ languages), **M**ulti-granularities (input length up to 8192), **M**ulti-Functionality (unification of dense, lexical, multi-vec/colbert retrieval).
41
+ It is the first embedding model which supports all three retrieval methods, achieving new SOTA on multi-lingual (MIRACL) and cross-lingual (MKQA) benchmarks.
42
+ [Technical Report](https://github.com/FlagOpen/FlagEmbedding/blob/master/FlagEmbedding/BGE_M3/BGE_M3.pdf) and [Code](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/BGE_M3). :fire:
43
+ - 1/9/2024: Release [Activation-Beacon](https://github.com/FlagOpen/FlagEmbedding/tree/master/Long_LLM/activation_beacon), an effective, efficient, compatible, and low-cost (training) method to extend the context length of LLM. [Technical Report](https://arxiv.org/abs/2401.03462) :fire:
44
+ - 12/24/2023: Release **LLaRA**, a LLaMA-7B based dense retriever, leading to state-of-the-art performances on MS MARCO and BEIR. Model and code will be open-sourced. Please stay tuned. [Technical Report](https://arxiv.org/abs/2312.15503) :fire:
45
+ - 11/23/2023: Release [LM-Cocktail](https://github.com/FlagOpen/FlagEmbedding/tree/master/LM_Cocktail), a method to maintain general capabilities during fine-tuning by merging multiple language models. [Technical Report](https://arxiv.org/abs/2311.13534) :fire:
46
+ - 10/12/2023: Release [LLM-Embedder](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/llm_embedder), a unified embedding model to support diverse retrieval augmentation needs for LLMs. [Technical Report](https://arxiv.org/pdf/2310.07554.pdf)
47
+ - 09/15/2023: The [technical report](https://arxiv.org/pdf/2309.07597.pdf) and [massive training data](https://data.baai.ac.cn/details/BAAI-MTP) of BGE has been released
48
  - 09/12/2023: New models:
49
  - **New reranker model**: release cross-encoder models `BAAI/bge-reranker-base` and `BAAI/bge-reranker-large`, which are more powerful than embedding model. We recommend to use/fine-tune them to re-rank top-k documents returned by embedding models.
50
  - **update embedding model**: release `bge-*-v1.5` embedding model to alleviate the issue of the similarity distribution, and enhance its retrieval ability without instruction.
 
69
 
70
  | Model | Language | | Description | query instruction for retrieval [1] |
71
  |:-------------------------------|:--------:| :--------:| :--------:|:--------:|
72
+ | [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) | Multilingual | [Inference](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/BGE_M3#usage) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/FlagEmbedding/BGE_M3) | Multi-Functionality(dense retrieval, sparse retrieval, multi-vector(colbert)), Multi-Linguality, and Multi-Granularity(8192 tokens) | |
73
  | [BAAI/llm-embedder](https://huggingface.co/BAAI/llm-embedder) | English | [Inference](./FlagEmbedding/llm_embedder/README.md) [Fine-tune](./FlagEmbedding/llm_embedder/README.md) | a unified embedding model to support diverse retrieval augmentation needs for LLMs | See [README](./FlagEmbedding/llm_embedder/README.md) |
74
  | [BAAI/bge-reranker-large](https://huggingface.co/BAAI/bge-reranker-large) | Chinese and English | [Inference](#usage-for-reranker) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/reranker) | a cross-encoder model which is more accurate but less efficient [2] | |
75
  | [BAAI/bge-reranker-base](https://huggingface.co/BAAI/bge-reranker-base) | Chinese and English | [Inference](#usage-for-reranker) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/reranker) | a cross-encoder model which is more accurate but less efficient [2] | |
 
86
  | [BAAI/bge-base-zh](https://huggingface.co/BAAI/bge-base-zh) | Chinese | [Inference](#usage-for-embedding-model) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/finetune) | a base-scale model but with similar ability to `bge-large-zh` | `为这个句子生成表示以用于检索相关文章:` |
87
  | [BAAI/bge-small-zh](https://huggingface.co/BAAI/bge-small-zh) | Chinese | [Inference](#usage-for-embedding-model) [Fine-tune](https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/finetune) | a small-scale model but with competitive performance | `为这个句子生成表示以用于检索相关文章:` |
88
 
 
89
  [1\]: If you need to search the relevant passages to a query, we suggest to add the instruction to the query; in other cases, no instruction is needed, just use the original query directly. In all cases, **no instruction** needs to be added to passages.
90
 
91
  [2\]: Different from embedding model, reranker uses question and document as input and directly output similarity instead of embedding. To balance the accuracy and time cost, cross-encoder is widely used to re-rank top-k documents retrieved by other simple models.