Gte Base En V1.5

由 Alibaba-NLP 开发

GTE-base-en-v1.5 是一个英文句子转换器模型，专注于句子相似度任务，在多个文本嵌入基准测试中表现优异。

文本嵌入

Transformers

支持多种语言

开源协议:Apache-2.0 #英文文本嵌入 #高精度相似度计算 #多任务评估优化

下载量 1.5M

发布时间 : 4/20/2024

模型介绍

内容详情

替代品

模型简介

该模型是一个通用的英文文本嵌入模型，能够将句子转换为高维向量表示，用于计算句子相似度、信息检索和文本分类等任务。

模型特点

高性能句子嵌入

在多个MTEB基准测试中表现出色，特别是在句子相似度和分类任务上

多任务适应能力

能够处理多种文本相关任务，包括分类、聚类、检索和重排序

开源许可

采用Apache-2.0许可证，允许商业和研究使用

模型能力

句子相似度计算

文本分类

信息检索

文本聚类

语义搜索

句子嵌入生成

使用案例

电子商务

产品评论分类

对亚马逊产品评论进行情感分析

在MTEB AmazonPolarityClassification上达到93.02%准确率

反事实分析

识别亚马逊产品评论中的反事实陈述

在MTEB AmazonCounterfactualClassification上达到74.79%准确率

金融

银行客服分类

对银行客户查询进行分类

在MTEB Banking77Classification上达到86.73%准确率

学术研究

论文聚类

对arXiv和biorxiv论文进行主题聚类

在MTEB ArxivClusteringP2P上达到47.51 V-measure

library_name: transformers tags:

sentence-transformers
gte
mteb
transformers.js
sentence-similarity license: apache-2.0 language:
en model-index:
name: gte-base-en-v1.5 results:
- task: type: Classification dataset: type: mteb/amazon_counterfactual name: MTEB AmazonCounterfactualClassification (en) config: en split: test revision: e8379541af4e31359cca9fbcf4b00f2671dba205 metrics:
  - type: accuracy value: 74.7910447761194
  - type: ap value: 37.053785713650626
  - type: f1 value: 68.51101510998551
- task: type: Classification dataset: type: mteb/amazon_polarity name: MTEB AmazonPolarityClassification config: default split: test revision: e2d317d38cd51312af73b3d32a06d1a08b442046 metrics:
  - type: accuracy value: 93.016875
  - type: ap value: 89.17750268426342
  - type: f1 value: 92.9970977240524
- task: type: Classification dataset: type: mteb/amazon_reviews_multi name: MTEB AmazonReviewsClassification (en) config: en split: test revision: 1399c76144fd37290681b995c656ef9b2e06e26d metrics:
  - type: accuracy value: 53.312000000000005
  - type: f1 value: 52.98175784163017
- task: type: Retrieval dataset: type: mteb/arguana name: MTEB ArguAna config: default split: test revision: c22ab2a51041ffd869aaddef7af8d8215647e41a metrics:
  - type: map_at_1 value: 38.193
  - type: map_at_10 value: 54.848
  - type: map_at_100 value: 55.388000000000005
  - type: map_at_1000 value: 55.388999999999996
  - type: map_at_3 value: 50.427
  - type: map_at_5 value: 53.105000000000004
  - type: mrr_at_1 value: 39.047
  - type: mrr_at_10 value: 55.153
  - type: mrr_at_100 value: 55.686
  - type: mrr_at_1000 value: 55.688
  - type: mrr_at_3 value: 50.676
  - type: mrr_at_5 value: 53.417
  - type: ndcg_at_1 value: 38.193
  - type: ndcg_at_10 value: 63.486
  - type: ndcg_at_100 value: 65.58
  - type: ndcg_at_1000 value: 65.61
  - type: ndcg_at_3 value: 54.494
  - type: ndcg_at_5 value: 59.339
  - type: precision_at_1 value: 38.193
  - type: precision_at_10 value: 9.075
  - type: precision_at_100 value: 0.9939999999999999
  - type: precision_at_1000 value: 0.1
  - type: precision_at_3 value: 22.096
  - type: precision_at_5 value: 15.619
  - type: recall_at_1 value: 38.193
  - type: recall_at_10 value: 90.754
  - type: recall_at_100 value: 99.431
  - type: recall_at_1000 value: 99.644
  - type: recall_at_3 value: 66.28699999999999
  - type: recall_at_5 value: 78.094
- task: type: Clustering dataset: type: mteb/arxiv-clustering-p2p name: MTEB ArxivClusteringP2P config: default split: test revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d metrics:
  - type: v_measure value: 47.508221208908964
- task: type: Clustering dataset: type: mteb/arxiv-clustering-s2s name: MTEB ArxivClusteringS2S config: default split: test revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53 metrics:
  - type: v_measure value: 42.04668382560096
- task: type: Reranking dataset: type: mteb/askubuntudupquestions-reranking name: MTEB AskUbuntuDupQuestions config: default split: test revision: 2000358ca161889fa9c082cb41daa8dcfb161a54 metrics:
  - type: map value: 61.828759903716815
  - type: mrr value: 74.37343358395991
- task: type: STS dataset: type: mteb/biosses-sts name: MTEB BIOSSES config: default split: test revision: d3fb88f8f02e40887cd149695127462bbcf29b4a metrics:
  - type: cos_sim_pearson value: 85.03673698773017
  - type: cos_sim_spearman value: 83.6470866785058
  - type: euclidean_pearson value: 82.64048673096565
  - type: euclidean_spearman value: 83.63142367101115
  - type: manhattan_pearson value: 82.71493099760228
  - type: manhattan_spearman value: 83.60491704294326
- task: type: Classification dataset: type: mteb/banking77 name: MTEB Banking77Classification config: default split: test revision: 0fd18e25b25c072e09e0d92ab615fda904d66300 metrics:
  - type: accuracy value: 86.73376623376623
  - type: f1 value: 86.70294049278262
- task: type: Clustering dataset: type: mteb/biorxiv-clustering-p2p name: MTEB BiorxivClusteringP2P config: default split: test revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40 metrics:
  - type: v_measure value: 40.31923804167062
- task: type: Clustering dataset: type: mteb/biorxiv-clustering-s2s name: MTEB BiorxivClusteringS2S config: default split: test revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908 metrics:
  - type: v_measure value: 37.552547125348454
- task: type: Retrieval dataset: type: mteb/cqadupstack-android name: MTEB CQADupstackAndroidRetrieval config: default split: test revision: f46a197baaae43b4f621051089b82a364682dfeb metrics:
  - type: map_at_1 value: 30.567
  - type: map_at_10 value: 41.269
  - type: map_at_100 value: 42.689
  - type: map_at_1000 value: 42.84
  - type: map_at_3 value: 37.567
  - type: map_at_5 value: 39.706
  - type: mrr_at_1 value: 37.053000000000004
  - type: mrr_at_10 value: 46.900999999999996
  - type: mrr_at_100 value: 47.662
  - type: mrr_at_1000 value: 47.713
  - type: mrr_at_3 value: 43.801
  - type: mrr_at_5 value: 45.689
  - type: ndcg_at_1 value: 37.053000000000004
  - type: ndcg_at_10 value: 47.73
  - type: ndcg_at_100 value: 53.128
  - type: ndcg_at_1000 value: 55.300000000000004
  - type: ndcg_at_3 value: 42.046
  - type: ndcg_at_5 value: 44.782
  - type: precision_at_1 value: 37.053000000000004
  - type: precision_at_10 value: 9.142
  - type: precision_at_100 value: 1.485
  - type: precision_at_1000 value: 0.197
  - type: precision_at_3 value: 20.076
  - type: precision_at_5 value: 14.535
  - type: recall_at_1 value: 30.567
  - type: recall_at_10 value: 60.602999999999994
  - type: recall_at_100 value: 83.22800000000001
  - type: recall_at_1000 value: 96.696
  - type: recall_at_3 value: 44.336999999999996
  - type: recall_at_5 value: 51.949
- task: type: Retrieval dataset: type: mteb/cqadupstack-english name: MTEB CQADupstackEnglishRetrieval config: default split: test revision: ad9991cb51e31e31e430383c75ffb2885547b5f0 metrics:
  - type: map_at_1 value: 28.538000000000004
  - type: map_at_10 value: 38.757999999999996
  - type: map_at_100 value: 40.129
  - type: map_at_1000 value: 40.262
  - type: map_at_3 value: 35.866
  - type: map_at_5 value: 37.417
  - type: mrr_at_1 value: 36.051
  - type: mrr_at_10 value: 44.868
  - type: mrr_at_100 value: 45.568999999999996
  - type: mrr_at_1000 value: 45.615
  - type: mrr_at_3 value: 42.558
  - type: mrr_at_5 value: 43.883
  - type: ndcg_at_1 value: 36.051
  - type: ndcg_at_10 value: 44.584
  - type: ndcg_at_100 value: 49.356
  - type: ndcg_at_1000 value: 51.39
  - type: ndcg_at_3 value: 40.389
  - type: ndcg_at_5 value: 42.14
  - type: precision_at_1 value: 36.051
  - type: precision_at_10 value: 8.446
  - type: precision