Search Results for author:
Found 1 papers, 0 papers with code
Defending Multimodal Fusion Models against Single-Source Adversaries
Beyond achieving high performance across many vision tasks, multimodal models are expected to be robust to single-source faults due to the availability of redundant information between modalities.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
