Susmit Jha

A new method enhances detection of backdoor threats in models using images and text.

Evaluating model calibration to improve code quality and developer trust.

Introducing TABDet, a new method for detecting backdoor attacks across NLP tasks.

A new method to verify deep neural networks using vision-language models.

A look into new methods for enhancing trust in AI responses.

Drones enhance search and rescue operations through smart decision-making and efficient planning.