Reproduce “Uncovering Semantic Selectivity of Latent Groups in Higher Visual Cortex with Mutual Information-Guided Diffusion” 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
Yule Wang, Joseph Yu, Chengrui Li, Weihan Li, Anqi Wu
MIG-Vis probes macaque IT responses by grouping them and guiding diffusion edits after one group is perturbed. Rotation- and category-informed groups receive supervision, while two groups are unsupervised. Edits associate one group with pose changes and another with category changes. Without mutual-information guidance, category transitions were less consistent in examples; synthesized images do not establish neural causality or universal coding.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.