Guangzhi Sun

A new dataset to aid recognition of audio and visual content in lectures.

Examining LLMs' potential in enhancing neuroscience research predictions.

CrossCheckGPT provides a new way to evaluate model reliability and accuracy.

A new method blends weak and strong AI models to align with human values.

Understanding how persona shapes interactions with conversational agents in various fields.

A novel method improves voice recognition accuracy across multiple languages.

This study addresses challenges in audio language models for low-resource languages.