# Adversarial Robustness

Polite Guard is an open-source NLP language model developed by Intel, based on BERT fine-tuned for text classification tasks, designed to classify text into four categories: polite, somewhat polite, neutral, and impolite.

FLODA is an advanced deepfake detection model that integrates image caption generation and authenticity assessment functions, achieving high-precision detection through visual question answering tasks.

Vision-language model initialized with OpenAI CLIP, enhanced robustness through unsupervised adversarial fine-tuning

TeCoA is a vision-language model initialized from OpenAI CLIP, enhanced with supervised adversarial fine-tuning for improved robustness

A question-answering model trained on the adversarial_qa dataset, designed to answer adversarial questions.