Previous works have implicitly studied that FL suffers from the "client-drift" problem, which is caused by the inconsistent optimum across local clients.

This work proposes a middle ground in the form of a scalable information-theoretic generalization of CCA, termed max-sliced mutual information (mSMI).

Competing objectives arise when a model's capabilities

In practice, multiple conflicting objectives and costly evaluations (e.g., wet-lab experiments) make the diversity of candidates

AudioSet) has an even larger negative impact.

IMP-MoE-L focusing on video tasks that achieves new state-of-the-art in zero-shot video classification: 77.0% on Kinetics-400, 76.8% on Kinetics-600, and 68.3% on