Jianyi Zhang

A look at the security threats posed by instruction-tuned Code LLMs.

A new framework combines federated learning and multimodal models to improve privacy and performance.

A new method enhances the accuracy of AI-generated responses.

New methods help machines find key information from spoken content.