Snowflake Arctic Embed v2.0 Open-Source Multilingual Model - Extracting Text Features and Calculating Similarities for Over 100 Languages

Snowflake Arctic Embed L V2.0

Developed by Snowflake

Snowflake Arctic Embed v2.0 is a multilingual sentence embedding model that supports text feature extraction and sentence similarity calculation for over 100 languages.

Text Embedding

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Multilingual Embedding #Sentence Similarity Calculation #High-dimensional Feature Extraction

Downloads 231.00k

Release Time : 11/8/2024

Model Overview

This model focuses on sentence-level feature extraction and similarity calculation, suitable for text processing tasks in multilingual environments.

Model Features

Multilingual Support

Supports text processing for over 100 languages, including major and some minor languages.

High-performance Sentence Embedding

Performs excellently in MTEB benchmark tests, suitable for various NLP tasks.

Versatile Applications

Can be used for various text processing tasks such as sentence similarity, classification, clustering, and retrieval.

Model Capabilities

Sentence feature extraction

Text similarity calculation

Multilingual text processing

Text classification

Text clustering

Information retrieval

Use Cases

E-commerce

Product Review Classification

Sentiment analysis and classification of Amazon product reviews

Achieved 74.26% accuracy in the MTEB Amazon Polarity Classification task

Counterfactual Review Detection

Identifying counterfactual reviews on the Amazon platform

Achieved 67.04% accuracy in the MTEB Amazon Counterfactual Classification task

Academic Research

Paper Clustering

Topic clustering of arXiv and BioRxiv papers

Achieved a V-measure of 43.97 in the MTEB ArxivClusteringP2P task

Q&A Systems

Q&A Retrieval

Question retrieval on technical Q&A platforms like CQADupstack

Achieved NDCG@10 of 55.80 in the MTEB CQADupstackAndroidRetrieval task

🚀 snowflake-arctic-embed-l-v2.0

This is a model for sentence similarity tasks, supporting multiple languages. It has achieved certain results in various datasets of MTEB, including classification, retrieval, clustering, reranking, and STS tasks.

📚 Documentation

Model Information

Property	Details
Pipeline Tag	sentence-similarity
Tags	sentence-transformers, feature-extraction, sentence-similarity, mteb, arctic, snowflake-arctic-embed, transformers.js
Languages Supported	af, ar, az, be, bg, bn, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, gl, gu, he, hi, hr, ht, hu, hy, id, is, it, ja, jv, ka, kk, km, kn, ko, ky, lo, lt, lv, mk, ml, mn, mr, ms, my, ne, nl, pa, pl, pt, qu, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, th, tl, tr, uk, ur, vi, yo, zh
Model Name	snowflake-arctic-embed-l-v2.0
License	apache-2.0

Results on Datasets

MTEB AmazonCounterfactualClassification (en-ext)

Dataset: MTEB AmazonCounterfactualClassification (en-ext)
Task Type: Classification
Metrics:
- Accuracy: 67.039
- F1: 55.1806
- F1 Weighted: 73.41149999999999
- AP: 17.9914
- AP Weighted: 17.9914
- Main Score: 67.039

MTEB AmazonCounterfactualClassification (en)

Dataset: MTEB AmazonCounterfactualClassification (en)
Task Type: Classification
Metrics:
- Accuracy: 65.59700000000001
- F1: 60.244299999999996
- F1 Weighted: 68.9975
- AP: 29.762100000000004
- AP Weighted: 29.762100000000004
- Main Score: 65.59700000000001

MTEB AmazonPolarityClassification (default)

Dataset: MTEB AmazonPolarityClassification (default)
Task Type: Classification
Metrics:
- Accuracy: 74.2565
- F1: 74.0291
- F1 Weighted: 74.0291
- AP: 68.7595
- AP Weighted: 68.7595
- Main Score: 74.2565

MTEB AmazonReviewsClassification (en)

Dataset: MTEB AmazonReviewsClassification (en)
Task Type: Classification
Metrics:
- Accuracy: 34.946
- F1: 34.2853
- F1 Weighted: 34.2853
- Main Score: 34.946

MTEB ArguAna (default)

Dataset: MTEB ArguAna (default)
Task Type: Retrieval
Metrics: A large number of metrics are provided, including ndcg_at_1, ndcg_at_3, etc. For details, please refer to the original document.
Main Score: 59.146

MTEB ArxivClusteringP2P (default)

Dataset: MTEB ArxivClusteringP2P (default)
Task Type: Clustering
Metrics:
- V Measure: 43.9715
- V Measure Std: 13.4325
- Main Score: 43.9715

MTEB ArxivClusteringS2S (default)

Dataset: MTEB ArxivClusteringS2S (default)
Task Type: Clustering
Metrics:
- V Measure: 34.775800000000004
- V Measure Std: 13.922799999999999
- Main Score: 34.775800000000004

MTEB AskUbuntuDupQuestions (default)

Dataset: MTEB AskUbuntuDupQuestions (default)
Task Type: Reranking
Metrics:
- MAP: 63.3521
- MRR: 77.5965
- nAUC_map_max: 21.2353
- nAUC_map_std: 17.002100000000002
- nAUC_map_diff1: 3.8135000000000003
- nAUC_mrr_max: 35.058299999999996
- nAUC_mrr_std: 20.432
- nAUC_mrr_diff1: 9.2584
- Main Score: 63.3521

MTEB BIOSSES (default)

Dataset: MTEB BIOSSES (default)
Task Type: STS
Metrics:
- Pearson: 89.8072
- Spearman: 87.2875
- Cosine Pearson: 89.8072
- Cosine Spearman: 87.2875
- Manhattan Pearson: 87.9173
- Manhattan Spearman: 86.7327
- Euclidean Pearson: 88.21600000000001
- Euclidean Spearman: 87.2875
- Main Score: 87.2875

MTEB Banking77Classification (default)

Dataset: MTEB Banking77Classification (default)
Task Type: Classification
Metrics:
- Accuracy: 81.8149
- F1: 81.2226
- F1 Weighted: 81.2226
- Main Score: To be continued (original document is incomplete)

📄 License

This project is licensed under the Apache-2.0 license.

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