M

Mdeberta V3 Base Mnli Xnli

Developed by MoritzLaurer
A multilingual natural language inference model supporting 100 languages, suitable for zero-shot classification tasks
Downloads 604.03k
Release Time : 3/2/2022

Model Overview

This model is based on the mDeBERTa-v3-base architecture, fine-tuned on XNLI and MNLI datasets, capable of handling multilingual zero-shot classification and natural language inference tasks

Model Features

Multilingual Support
Supports zero-shot classification and natural language inference in 100 languages
High Performance
Achieves an average accuracy of 80.8% on the XNLI test set, making it one of the best-performing multilingual base-size Transformer models
Zero-shot Classification Capability
Performs classification tasks without task-specific training data
Cross-lingual Transfer
Capable of performing NLI tasks on languages without specific training data

Model Capabilities

Multilingual text classification
Natural language inference
Zero-shot learning
Cross-lingual transfer

Use Cases

Text Classification
News Classification
Classify news texts into predefined categories (e.g., politics, economy, entertainment)
Content Moderation
Multilingual Content Classification
Classify and moderate multilingual user-generated content
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase