Qihang(Lucas) Jin 金启航

Developing and optimizing vision-language models for multimodal understanding, exploring representation learning and fusion strategies for multimodal data, focusing on efficient training strategies and robust performance across diverse domains.

Researching efficient fine-tuning methods, parameter-efficient techniques, and optimization strategies to reduce computational costs while maintaining model performance.

Addressing critical challenges in AI deployment including safety, privacy protection, bias mitigation, and responsible AI development across real-world applications.

Bridging AI and healthcare through interdisciplinary approaches, integrating neuroscience, computer vision, and clinical expertise for brain-computer interfaces, medical imaging, and intelligent healthcare systems.