Reproduce “Trusted Multi-View Classification with Dynamic Evidential Fusion” 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

Zongbo Han, Changqing Zhang, Huazhu Fu, Joey Tianyi Zhou

TMC combines multiple views by turning each view’s evidence and uncertainty into subjective opinions, then fusing them with Dempster–Shafer theory. ETMC adds a pseudo-view from concatenated features. Under Gaussian test noise, both generally retain accuracy better as noise rises; on NYUD Depth V2, ETMC reaches 72.5%, versus 70.9% for the strongest prior method. These results establish no broader robustness guarantee.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.