M

Modernbert Embed Base Legal MRL

Developed by AdamLucek
A legal domain sentence embedding model fine-tuned based on ModernBERT, supporting multi-level dimensional output, suitable for legal text similarity calculation and information retrieval tasks.
Downloads 40
Release Time : 1/20/2025

Model Overview

This is a sentence embedding model optimized for the legal domain, capable of converting text into 768-dimensional vectors, supporting multi-level dimensional output (768/512/256/128/64 dimensions), particularly suitable for semantic similarity calculation, information retrieval, and clustering analysis of legal documents.

Model Features

Multi-level Dimensional Output
Supports 768/512/256/128/64-dimensional multi-level embedding output, allowing flexible dimension selection based on application scenarios
Legal Domain Optimization
Fine-tuned with synthetic legal domain data, excelling in processing legal texts
Long Text Support
Supports sequences up to 8192 tokens, ideal for processing long texts such as legal documents
Efficient Retrieval Capability
Performs exceptionally well in information retrieval tasks, especially in legal document retrieval scenarios

Model Capabilities

Semantic text similarity calculation
Semantic search
Information retrieval
Text clustering
Feature extraction

Use Cases

Legal Document Processing
Legal Case Retrieval
Quickly retrieve legal documents related to the query case
Achieved a normalized discounted cumulative gain@10 of 0.63 on the test set
Contract Clause Matching
Identify similar clauses and related content in contracts
Information Retrieval Systems
Legal Q&A System
Build a semantic retrieval-based legal Q&A system
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase