MGTBench: Benchmarking Machine-Generated Text Detection
Paper
•
2303.14822
•
Published
Model Card for Model ID
This modelcard aims to be a base template for new models. It has been generated using this raw template.
Uses
This model is used for detecting Machine-generated Texts.
Training Details
Training Data
This model trains on a corpus of both human and StableLM-generated answers for questions from SQuAD1 dataset. The dataset can be found here.
Training Procedure
This model is created by fine-tuning Distilbert-base-uncased model. The training procedure follows He et al.'s instructions.
- Downloads last month
- 4