S

Splade Legal French

Developed by maastrichtlawtech
SPLADEv2 model based on Camembert architecture, optimized for French paragraph retrieval and sentence similarity tasks
Downloads 19
Release Time : 6/4/2024

Model Overview

This model is a dense retrieval model based on the SPLADEv2 architecture, using Camembert-base as the base model. It performs exceptionally well in French information retrieval tasks, particularly suitable for paragraph retrieval and sentence similarity calculations in the legal domain

Model Features

French optimization
Specifically optimized for French text, excelling in French information retrieval tasks
Legal domain applicability
Trained on the maastrichtlawtech/lleqa legal QA dataset, particularly suitable for legal document retrieval
Efficient retrieval
Utilizes SPLADEv2 architecture for efficient dense paragraph retrieval

Model Capabilities

Calculate sentence similarity
Perform paragraph retrieval
Process French text
Analyze legal documents

Use Cases

Legal information retrieval
Legal QA system
Retrieve relevant legal provisions and precedents in legal QA systems
High-recall legal document retrieval
Document similarity analysis
Compare similarities between legal documents
Accurately identify similar legal cases
General information retrieval
French search engine
Serve as the retrieval backend for French search engines
Efficient and accurate French content retrieval
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase