gte-Qwen2-7B-instruct is the latest model in the gte (General Text Embedding) model family that ranks No.1 in both English and Chinese evaluations on the Massive Text Embedding Benchmark MTEB benchmark (as of June 16, 2024).
Recently, the Qwen team released the Qwen2 series models, and we have trained the gte-Qwen2-7B-instruct model based on the Qwen2-7B LLM model. Compared to the gte-Qwen1.5-7B-instruct model, the gte-Qwen2-7B-instruct model uses the same training data and training strategies during the finetuning stage, with the only difference being the upgraded base model to Qwen2-7B. Considering the improvements in the Qwen2 series models compared to the Qwen1.5 series, we can also expect consistent performance enhancements in the embedding models.
The model incorporates several key advancements:
Integration of bidirectional attention mechanisms, enriching its contextual understanding.
Instruction tuning, applied solely on the query side for streamlined efficiency
Comprehensive training across a vast, multilingual text corpus spanning diverse domains and scenarios. This training leverages both weakly supervised and supervised data, ensuring the model's applicability across numerous languages and a wide array of downstream tasks.
Model Information
Model Size: 7B
Embedding Dimension: 3584
Max Input Tokens: 32k
Requirements
transformers>=4.39.2
flash_attn>=2.5.6
Usage
Sentence Transformers
python
1from sentence_transformers import SentenceTransformer
23model = SentenceTransformer("Alibaba-NLP/gte-Qwen2-7B-instruct", trust_remote_code=True)4# In case you want to reduce the maximum length:5model.max_seq_length =819267queries =[8"how much protein should a female eat",9"summit define",10]11documents =[12"As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",13"Definition of summit for English Language Learners. : 1 the highest point of a mountain : the top of a mountain. : 2 the highest level. : 3 a meeting or series of meetings between the leaders of two or more governments.",14]1516query_embeddings = model.encode(queries, prompt_name="query")17document_embeddings = model.encode(documents)1819scores =(query_embeddings @ document_embeddings.T)*10020print(scores.tolist())
Observe the config_sentence_transformers.json to see all pre-built prompt names. Otherwise, you can use model.encode(queries, prompt="Instruct: ...\nQuery: " to use a custom prompt of your choice.
Transformers
python
1import torch
2import torch.nn.functional as F
34from torch import Tensor
5from transformers import AutoTokenizer, AutoModel
678deflast_token_pool(last_hidden_states: Tensor,9 attention_mask: Tensor)-> Tensor:10 left_padding =(attention_mask[:,-1].sum()== attention_mask.shape[0])11if left_padding:12return last_hidden_states[:,-1]13else:14 sequence_lengths = attention_mask.sum(dim=1)-115 batch_size = last_hidden_states.shape[0]16return last_hidden_states[torch.arange(batch_size, device=last_hidden_states.device), sequence_lengths]171819defget_detailed_instruct(task_description:str, query:str)->str:20returnf'Instruct: {task_description}\nQuery: {query}'212223# Each query must come with a one-sentence instruction that describes the task24task ='Given a web search query, retrieve relevant passages that answer the query'25queries =[26 get_detailed_instruct(task,'how much protein should a female eat'),27 get_detailed_instruct(task,'summit define')28]29# No need to add instruction for retrieval documents30documents =[31"As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",32"Definition of summit for English Language Learners. : 1 the highest point of a mountain : the top of a mountain. : 2 the highest level. : 3 a meeting or series of meetings between the leaders of two or more governments."33]34input_texts = queries + documents
3536tokenizer = AutoTokenizer.from_pretrained('Alibaba-NLP/gte-Qwen2-7B-instruct', trust_remote_code=True)37model = AutoModel.from_pretrained('Alibaba-NLP/gte-Qwen2-7B-instruct', trust_remote_code=True)3839max_length =81924041# Tokenize the input texts42batch_dict = tokenizer(input_texts, max_length=max_length, padding=True, truncation=True, return_tensors='pt')43outputs = model(**batch_dict)44embeddings = last_token_pool(outputs.last_hidden_state, batch_dict['attention_mask'])4546# normalize embeddings47embeddings = F.normalize(embeddings, p=2, dim=1)48scores =(embeddings[:2] @ embeddings[2:].T)*10049print(scores.tolist())
Evaluation
MTEB & C-MTEB
You can use the scripts/eval_mteb.py to reproduce the following result of gte-Qwen2-7B-instruct on MTEB(English)/C-MTEB(Chinese):
The gte series models have consistently released two types of models: encoder-only models (based on the BERT architecture) and decode-only models (based on the LLM architecture).
If you find our paper or models helpful, please consider cite:
@article{li2023towards,
title={Towards general text embeddings with multi-stage contrastive learning},
author={Li, Zehan and Zhang, Xin and Zhang, Yanzhao and Long, Dingkun and Xie, Pengjun and Zhang, Meishan},
journal={arXiv preprint arXiv:2308.03281},
year={2023}
}
