Takumi Fukami

FedProx-based heterogeneity-aware parameter-free federated learning

Plausible Token Amplification for Improving Accuracy of Differentially Private In-Context Learning Based on Implicit Bayesian Inference