AIbase
EN
Home
Roberta Base RTE
R

Roberta Base RTE

Developed by textattack
Sequence classification model fine-tuned on the GLUE dataset using the TextAttack framework, based on the roberta-base architecture
Text Classification#GLUE fine-tuning#Text adversarial training#Sequence classification
Downloads 29
Release Time : 3/2/2022

Model Overview

This pre-trained model is designed for text sequence classification tasks, particularly suited for natural language understanding tasks. Fine-tuned using the adversarial training framework TextAttack, it enhances the model's robustness in classification tasks.

Model Features

Adversarial training enhancement
Adversarial training via the TextAttack framework improves the model's robustness against adversarial examples
GLUE dataset fine-tuning
Fine-tuned on the General Language Understanding Evaluation (GLUE) benchmark, suitable for various natural language understanding tasks
Efficient training configuration
Optimized with a learning rate of 2e-05 and a maximum sequence length of 128, achieving peak performance within 5 epochs

Model Capabilities

Text classification
Natural language understanding
Sequence labeling

Use Cases

Text analysis
Sentiment analysis
Classifies the sentiment orientation of text content
Achieved 79.42% accuracy on the validation set
Topic classification
Categorizes text into predefined topic classes
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase