updated 1 public sources
multimodal large language modelsvideo understandingaction recognitionlightweight networks

Current frame

PhD student at Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, researching multimodal large language models and video understanding.

Extended note

Public researcher profiles identify Kunchang Li as a PhD student at Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, after undergraduate study at Beihang University. His listed expertise includes multimodal large language models, video understanding, action recognition, and lightweight networks. ByteDance Seed and arXiv source pages also list him as an author on Seed multimodal research publications, including the Seed1.5-VL technical report.