X

Xiaobu Embedding

Developed by lier007
xiaobu-embedding is a multi-task embedding model that supports various Chinese natural language processing tasks, including text similarity calculation, classification, clustering, and retrieval.
Downloads 147
Release Time : 1/9/2024

Model Overview

This model focuses on learning embedding representations for Chinese texts, capable of generating high-quality text vectors suitable for various downstream tasks such as semantic similarity calculation, text classification, and information retrieval.

Model Features

Multi-task Support
Supports various natural language processing tasks, including text similarity calculation, classification, clustering, and retrieval.
Chinese Optimization
Specifically optimized for Chinese texts, better capturing Chinese semantic features.
High Performance
Outstanding performance on multiple Chinese benchmark tests, particularly excelling in medical-related tasks.

Model Capabilities

Text Embedding
Semantic Similarity Calculation
Text Classification
Text Clustering
Information Retrieval
Re-ranking

Use Cases

Medical Field
Medical QA Retrieval
Used for retrieval and answering of medical-related questions
Achieved MAP@10 of 37.604 on CMedQA retrieval task
Medical Document Re-ranking
Re-ranking medical-related documents by relevance
Achieved MAP of 87.57 on CMedQAv2 re-ranking task
E-commerce
Product Review Classification
Sentiment and topic classification of product reviews
Achieved accuracy of 86.74% on JD.com review classification task
Product Retrieval
Product search functionality in e-commerce platforms
Achieved MAP@10 of 63.14 on EcomRetrieval task
General NLP
Text Similarity Calculation
Calculating semantic similarity between two texts
Achieved Pearson correlation of 79.75 on STSB task
Text Classification
Multi-category classification of texts
Achieved accuracy of 49.74% on IFlyTek classification task
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase