gte-small Open-source Text Embedding Model - Freely Assist in Sentence Similarity Calculation and Text Classification Retrieval

Gte Small

Developed by thenlper

GTE-small is a compact general-purpose text embedding model suitable for various natural language processing tasks, including sentence similarity calculation, text classification, and retrieval.

Text Embedding EnglishOpen Source License:MIT #Sentence Similarity Calculation #Multi-task Text Embedding #High-precision Classification

Downloads 450.86k

Release Time : 7/27/2023

Model Overview

GTE-small is a text embedding model based on the sentence transformer architecture, primarily designed to generate high-quality sentence-level embeddings and supports multiple downstream NLP tasks.

Model Features

Multi-task Support

Supports various natural language processing tasks, including classification, retrieval, and clustering.

Efficient Performance

Demonstrates outstanding performance in multiple benchmarks, particularly in text classification tasks.

General Text Embedding

Capable of generating high-quality sentence-level embeddings suitable for diverse downstream applications.

Model Capabilities

Sentence similarity calculation

Text classification

Information retrieval

Text clustering

Semantic text similarity evaluation

Use Cases

E-commerce

Product Review Classification

Sentiment polarity classification for Amazon product reviews

Achieved 91.8% accuracy on the AmazonPolarity classification task

Counterfactual Review Identification

Identifying counterfactual reviews on the Amazon platform

Achieved 73.2% accuracy on the AmazonCounterfactual classification task

Academic Research

Paper Clustering

Topic clustering for arXiv and biorxiv papers

Achieved a V-measure of 47.9 on the arXiv paper clustering task

Q&A Systems

Duplicate Question Detection

Identifying duplicate questions on the AskUbuntu forum

Achieved an average precision of 61.7 in the re-ranking task

🚀 gte-small

This document presents the performance metrics of the gte-small model across various natural language processing tasks, including classification, retrieval, clustering, reranking, and semantic text similarity.

📚 Documentation

Model Information

Property	Details
Model Name	gte-small
Tags	mteb, sentence-similarity, sentence-transformers, Sentence Transformers

Performance Metrics

Classification Tasks

Dataset	Accuracy	AP	F1
MTEB AmazonCounterfactualClassification (en)	73.22388059701493	36.09895941426988	67.3205651539195
MTEB AmazonPolarityClassification	91.81894999999999	88.5240138417305	91.80367382706962
MTEB AmazonReviewsClassification (en)	48.032	-	47.4490665674719
MTEB Banking77Classification	84.0844155844156	-	84.01485017302213

Retrieval Tasks

Dataset	MAP@1	MAP@10	MAP@100	MAP@1000	MRR@1	MRR@10	MRR@100	MRR@1000	NDCG@1	NDCG@10	NDCG@100	NDCG@1000	Precision@1	Precision@10	Precision@100	Precision@1000	Recall@1	Recall@10	Recall@100	Recall@1000
MTEB ArguAna	30.725	46.604	47.535	47.538000000000004	31.223	46.794000000000004	47.725	47.727000000000004	30.725	55.440999999999995	59.134	59.199	30.725	8.364	0.991	0.1	30.725	83.64200000000001	99.14699999999999	99.644
MTEB CQADupstackAndroidRetrieval	30.261	42.419000000000004	43.927	44.055	36.91	48.02	48.658	48.708	36.91	49.353	54.456	56.363	36.91	9.700000000000001	1.557	0.202	30.261	63.242	84.09100000000001	96.143
MTEB CQADupstackEnglishRetrieval	31.145	40.996	42.266999999999996	42.397	38.344	46.827000000000005	47.446	47.489	38.344	46.733000000000004	51.103	53.075	38.344	8.822000000000001	1.417	0.187	31.145	56.909	75.274	87.629
MTEB CQADupstackGamingRetrieval	38.83	51.553000000000004	52.581	52.638	44.513999999999996	54.998000000000005	55.650999999999996	55.679	44.513999999999996	57.67400000000001	61.663999999999994	62.743	44.513999999999996	9.423	1.2309999999999999	0.13699999999999998	38.83	72.327	89.519	97.041
MTEB CQADupstackGisRetrieval	25.484	34.527	35.661	35.739	27.458	36.543	37.482	37.543	27.458	39.396	44.742	46.708	27.458	5.976999999999999	0.907	0.11100000000000002	25.484	52.317	76.701	91.408
MTEB CQADupstackMathematicaRetrieval	16.719	25.269000000000002	26.442	26.557	20.896	29.982999999999997	30.895	30.961	20.896	30.814000000000004	36.418	39.182	20.896	5.821	0.991	0.136	16.719	43.155	67.831	87.617

Clustering Tasks

Dataset	V-Measure
MTEB ArxivClusteringP2P	47.90178078197678
MTEB ArxivClusteringS2S	40.25728393431922
MTEB BiorxivClusteringP2P	38.36574769259432
MTEB BiorxivClusteringS2S	35.4857033165287

Reranking Task

Dataset	MAP	MRR
MTEB AskUbuntuDupQuestions	61.720297062897764	75.24139295607439

Semantic Textual Similarity (STS) Task

Dataset	Cosine Similarity Pearson	Cosine Similarity Spearman	Euclidean Pearson	Euclidean Spearman	Manhattan Pearson	Manhattan Spearman
MTEB BIOSSES	89.43527309184616	88.17128615100206	87.89922623089282	87.96104039655451	87.9818290932077	88.00923426576885

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご