Reproduce “MouseGPT: A Large-scale Vision-Language Model for Mouse Behavior Analysis” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Teng Xu, Taotao Zhou, Youjia Wang, Peng Yang, Simin Tang, Kuixiang Shao, Zifeng Tang, Yifei Liu and 10 more
MouseGPT turns mouse video into natural-language descriptions of actions and body-part postures, then uses text embeddings as features for behavioral analysis. In drug experiments, posture proportions and timing changed within walking, while walking subtype categories did not; this finding is limited to the reported mouse experiments. Its training descriptions were judged factually correct in 78.6% of cases on average by an expert panel.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.