Cloudy-large-zh Open-source Model - Easily Calculate Chinese Sentence Similarities, Support Multi-task Retrieval and Sorting

Cloudy Large Zh

Developed by akarum

cloudy-large-zh is a Chinese sentence similarity calculation model, focusing on feature extraction and sentence similarity tasks, supporting various retrieval and re-ranking tasks.

Text Embedding

Safetensors

#Chinese Sentence Similarity #Medical Retrieval Optimization #High-Dimensional Feature Extraction

Downloads 56

Release Time : 12/25/2023

Model Overview

This model is primarily used for Chinese text sentence similarity calculation and feature extraction, suitable for various information retrieval and re-ranking scenarios, especially excelling in medical and e-commerce fields.

Model Features

High-Performance Chinese Sentence Similarity Calculation

Performs exceptionally well on multiple Chinese datasets, particularly in retrieval tasks within the medical and e-commerce fields.

Multi-Task Support

Supports various tasks such as sentence similarity, feature extraction, re-ranking, and retrieval.

Instruction-Free

Does not require additional instructions in retrieval tasks, simplifying the usage process.

Model Capabilities

Sentence similarity calculation

Feature extraction

Re-ranking

Information retrieval

Use Cases

Medical Field

Medical Q&A Retrieval

Used for retrieval and re-ranking tasks in medical Q&A systems to improve answer accuracy and relevance.

Achieved average accuracy rates of 86.10% and 86.95% on the CMedQAv1 and CMedQAv2 datasets, respectively.

COVID-19 Information Retrieval

Used for retrieving COVID-19-related information to quickly find the most relevant medical details.

Achieved a top-1 average accuracy rate of 74.76% and a top-10 average accuracy rate of 82.60%.

E-commerce Field

Product Retrieval

Used for product retrieval on e-commerce platforms to enhance user search experience.

Achieved a top-1 average accuracy rate of 52.30% and a top-10 average accuracy rate of 62.15%.

Video Retrieval

Video Content Retrieval

Used for video content retrieval and similarity calculation.

Achieved a top-1 average accuracy rate of 59.50% and a top-10 average accuracy rate of 69.53%.

🚀 Cloudy-Large-ZH: Sentence Similarity Model

This is a cloudy-large-zh model designed for sentence similarity tasks. It belongs to the category of sentence transformers and can be used for feature extraction and various retrieval and reranking tasks.

📚 Documentation

Model Information

Property	Details
Pipeline Tag	Sentence Similarity
Tags	Sentence Transformers, Feature Extraction, Sentence Similarity, MTEB
Model Name	cloudy-large-zh

Performance Metrics

Reranking Tasks

Task	Dataset	MAP	MRR
Reranking	MTEB CMedQAv1	86.10362876754219	88.77880952380951
Reranking	MTEB CMedQAv2	86.94664825874587	89.47257936507937
Reranking	MTEB MMarcoReranking	24.260352944026806	22.69484126984127
Reranking	MTEB T2Reranking	66.83154421352779	76.27995669041708

Retrieval Tasks

Task	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
Retrieval	MTEB CmedqaRetrieval	25.296999999999997	37.159	39.016	39.134	38.435	46.235	47.265	47.308	38.435	43.578	50.995000000000005	53.012	38.435	9.607000000000001	1.557	0.182	25.296999999999997	53.408	84.202	97.61
Retrieval	MTEB CovidRetrieval	74.763	82.604	82.795	82.798	74.816	82.601	82.787	82.78999999999999	74.921	85.83	86.655	86.748	74.921	9.663	1.0030000000000001	0.101	74.763	95.627	99.262	100.0
Retrieval	MTEB DuRetrieval	26.021	78.561	81.291	81.34400000000001	89.8	92.905	92.976	92.979	89.8	86.203	88.955	89.442	89.8	41.175	4.744000000000001	0.486	26.021	87.25	96.154	98.615
Retrieval	MTEB EcomRetrieval	52.300000000000004	62.149	62.719	62.73	52.300000000000004	62.149	62.719	62.73	52.300000000000004	66.99300000000001	69.672	69.95400000000001	52.300000000000004	8.219999999999999	0.9450000000000001	0.097	52.300000000000004	82.19999999999999	94.5	96.7
Retrieval	MTEB MMarcoRetrieval	64.888	73.921	74.28099999999999	74.295	67.092	74.547	74.862	74.875	67.092	77.687	79.24600000000001	79.60000000000001	67.092	9.424000000000001	1.019	0.105	64.888	88.672	95.599	98.337
Retrieval	MTEB MedicalRetrieval	55.50000000000001	61.316	61.832	61.867000000000004	55.7	61.416000000000004	61.931999999999995	61.967000000000006	55.50000000000001	64.228	67.04599999999999	68.176	55.50000000000001	7.340000000000001	0.873	0.097	55.50000000000001	73.4	87.3	96.6
Retrieval	MTEB T2Retrieval	28.303	76.943	80.585	80.657	90.742	93.496	93.55799999999999	93.56	90.742	84.94	88.616	89.25	90.742	41.507	4.984999999999999	0.515	28.303	83.895	95.537	98.558
Retrieval	MTEB VideoRetrieval	59.5	69.53	69.976	69.99300000000001	59.5	69.53	69.976	69.99300000000001	59.5	-	-	-	59.5	-	-	-	59.5	-	-	-

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