Junjie Fei

Junjie Fei is an active researcher in Computer Science with a focus on Multimodal Learning, Vision-Language Understanding, and Retrieval-Augmented Generation. They have collaborated with prominent researchers like Mohamed Elhoseiny, Sanjoy Chowdhury, and Jun Chen on various projects.

Key research contributions include:

• Part-Aware 3D Vision-Language Understanding (2024) with novel tasks like Part-Aware Point Grounding
• MAGNET (2025) - a multi-agent framework for audio-visual retrieval
• WikiAutoGen (2025) - automated Wikipedia-style multimodal article generation
• V-RAG (2024) - vision-centric retrieval-augmented generation for document haystacks

While no explicit academic affiliation or awards are stated in the provided text, their work spans 3D object segmentation, multi-video temporal grounding, multimodal knowledge integration, and large document reasoning, indicating significant contributions to AI research.