J

Japanese Bge Reranker V2 M3 V1

Developed by hotchpotch
This is a Japanese Reranker (Cross-Encoder) model for text ranking tasks, featuring 24 layers and a hidden layer size of 1024.
Downloads 1,151
Release Time : 3/28/2024

Model Overview

This model is part of the Japanese Reranker series, specifically designed for Japanese text ranking tasks, capable of scoring the relevance between queries and passages.

Model Features

High-Performance Japanese Reranker
Excels on multiple Japanese datasets, outperforming other similar models.
Multi-Dataset Training
Trained on multiple datasets including JQaRA, JGLUE, MIRACL, mr-tydi, and mmarco.
Multiple Size Options
Offers models ranging from xsmall to large to accommodate different computational resource needs.

Model Capabilities

Japanese Text Ranking
Query-Passage Relevance Scoring
Cross-Language Text Ranking

Use Cases

Information Retrieval
Search Engine Result Ranking
Used to improve the relevance ranking of Japanese query results in search engines.
Achieved a score of 0.6918 on the JQaRA dataset.
Question Answering Systems
Answer Ranking in QA Systems
Used to rank candidate answers by relevance in question answering systems.
Achieved a score of 0.9624 on the JSQuAD dataset.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase