daalft/xlmr-ai-text-detection

Overview

This model fine-tunes the XLM-RoBERTa-base model with the train partition (346977 unique entries) of the Kaggle dataset Augmented data for LLM - Detect AI Generated Text. The training data is English, but the model should have cross-lingual capabilities. The model has not been fine-tuned to reach state-of-the-art, but serves more as a proof-of-concept for low-resource languages.

Task

The model is trained to predict whether a text/essay is written by a human (output label 0) or AI (output label 1).