R

Ruri Reranker Stage1 Large

Developed by cl-nagoya
Ruri-Reranker is a Japanese text re-ranking model based on sentence-transformers, specifically designed to optimize the relevance ranking between queries and documents.
Downloads 23
Release Time : 8/19/2024

Model Overview

This model is primarily used for Japanese text re-ranking tasks, capable of evaluating the relevance between queries and documents to optimize search results. Built on the cl-nagoya/ruri-pt-large base model, it supports sequences up to 512 tokens in length.

Model Features

High-performance Japanese Re-ranking
Performs exceptionally well on multiple Japanese benchmarks (JQaRA, JaCWIR, MIRACL), surpassing similar models.
Large-capacity Model
With 337M parameters (303M excluding embeddings), it possesses powerful semantic understanding capabilities.
Long Text Support
Supports sequences up to 512 tokens in length, suitable for processing longer queries and documents.

Model Capabilities

Query-Document Relevance Scoring
Search Result Optimization
Japanese Text Understanding

Use Cases

Information Retrieval
Search Engine Result Optimization
Re-ranks search engine results to improve the ranking of the most relevant results.
Achieved a score of 77.1 on the JQaRA benchmark
Q&A Systems
Evaluates the relevance of candidate answers to questions to select the best answer.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase