Snowflake Arctic Embed M Open-source Model - Efficiently Extract Text Features and Accurately Calculate Sentence Similarity

Snowflake Arctic Embed M

Developed by Snowflake

Snowflake Arctic Embed M is a sentence transformer model focused on sentence similarity tasks, capable of efficiently extracting text features and calculating similarity between sentences.

Text Embedding

Transformers

Open Source License:Apache-2.0 #Sentence Similarity Calculation #Multi-task Evaluation #High-precision Semantic Matching

Downloads 722.08k

Release Time : 4/11/2024

Model Overview

This model is primarily used for sentence similarity calculation and feature extraction, demonstrating excellent performance on multiple standard datasets and supporting various natural language processing tasks.

Model Features

Efficient Sentence Similarity Calculation

Performs excellently on multiple standard datasets, accurately calculating semantic similarity between sentences.

Multi-task Support

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

MTEB Benchmark Validation

Demonstrates outstanding performance in multiple MTEB benchmarks, validating the model's broad applicability.

Model Capabilities

Sentence Similarity Calculation

Text Feature Extraction

Text Classification

Text Clustering

Information Retrieval

Use Cases

E-commerce

Product Review Classification

Sentiment classification of Amazon product reviews

Achieved 82.83% accuracy on the AmazonPolarity dataset

Counterfactual Review Detection

Identifying counterfactual reviews on Amazon

Achieved 76.81% accuracy on the AmazonCounterfactual dataset

Finance

Bank Customer Service Question Classification

Automatic classification of bank customer inquiries

Achieved 80.34% accuracy on the Banking77 dataset

Academic Research

Paper Clustering

Topic clustering of arXiv and biorxiv papers

Achieved a V-measure of 47.24% on the ArxivClusteringP2P dataset

🚀 snowflake-arctic-embed-m

This is a model for sentence similarity tasks. It has been tested on multiple datasets, showing its performance in various natural language processing tasks such as classification, retrieval, clustering, and reranking.

📚 Documentation

Model Information

Property	Details
Pipeline Tag	sentence-similarity
Tags	sentence-transformers, feature-extraction, sentence-similarity, mteb, arctic, snowflake-arctic-embed, transformers.js
Model Name	snowflake-arctic-embed-m

Performance Metrics

Classification Tasks

Dataset	Accuracy	AP	F1
MTEB AmazonCounterfactualClassification (en)	76.80597014925374	39.31198155789558	70.48198448222148
MTEB AmazonPolarityClassification	82.831525	77.4474050181638	82.77204845110204
MTEB AmazonReviewsClassification (en)	38.93000000000001	-	37.98013371053459
MTEB Banking77Classification	80.33766233766232	-	79.64931422442245

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	31.223	47.43	48.208	48.211	31.65	47.573	48.359	48.362	31.223	56.436	59.657000000000004	59.731	31.223	8.527999999999999	0.991	0.1	31.223	85.277	99.075	99.644
MTEB CQADupstackAndroidRetrieval	34.079	46.35	47.785	47.903	41.345	52.032000000000004	52.690000000000005	52.727999999999994	41.345	53.027	57.962	59.611999999999995	41.345	10.157	1.567	0.199	34.079	65.93900000000001	86.42699999999999	96.61
MTEB CQADupstackEnglishRetrieval	33.289	43.681	45.056000000000004	45.171	41.146	49.604	50.28399999999999	50.322	41.146	49.43	54.01899999999999	55.803000000000004	41.146	9.268	1.4749999999999999	0.19	33.289	59.209999999999994	78.676	89.84100000000001
MTEB CQADupstackGamingRetrieval	44.483	56.862	57.901	57.948	50.658	60.281	60.946	60.967000000000006	50.658	62.339	66.28399999999999	67.166	50.658	9.762	1.26	0.13799999999999998	44.483	74.819	91.702	97.84
MTEB CQADupstackGisRetrieval	26.489	37.004999999999995	38.001000000000005	38.085	28.362	38.807	39.671	39.736	28.362	42.510999999999996	47.226	49.226	28.362	6.633	0.9490000000000001	0.11499999999999999	26.489	57.457	78.712	93.565
MTEB CQADupstackMathematicaRetrieval	12.418999999999999	22.866	24.365000000000002	24.479	14.677000000000001	26.316	27.514	27.57	14.677000000000001	-	-	-	14.677000000000001	4.476	0.594	0.067	12.418999999999999	33.15	53.46	68.4

Clustering Tasks

Dataset	V-Measure
MTEB ArxivClusteringP2P	47.23625429411296
MTEB ArxivClusteringS2S	37.433880471403654
MTEB BigPatentClustering	19.116028913890613
MTEB BiorxivClusteringP2P	36.966921852810174
MTEB BiorxivClusteringS2S	31.98019698537654

Reranking Task

Dataset	MAP	MRR
MTEB AskUbuntuDupQuestions	60.53175025582013	74.51160796728664

STS Task

Dataset	Cosine Similarity Pearson	Cosine Similarity Spearman	Euclidean Pearson	Euclidean Spearman	Manhattan Pearson	Manhattan Spearman
MTEB BIOSSES	88.93746103286769	86.62245567912619	87.154173907501	86.62245567912619	87.17682026633462	86.74775973908348

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