Xiaobao Wang

Achieving 90% accuracy in Alzheimer's detection through a novel graph-based approach that captures the intricacies of spontaneous speech could revolutionize diagnostic practices.

Iteratively prompting a graph neural network at test time to amplify out-of-distribution signals dramatically improves OOD detection accuracy.

Overcome Alzheimer's speech detection's data scarcity with FAL-AD, a federated learning framework that hits 91.52% accuracy by generating synthetic speech samples and aligning acoustic and textual features.