Different from the original MoCo [19] and its cross-modal versions [15, 33, 35] that utilize momentum update for only momentum encoders to maintain a large consistent queue, our BMU strategy imposes momentum update on both momentum encoders and (video/text) encoders.

We reveal which problems in unbalanced optimal transport can/cannot be solved efficiently.

We demonstrate this by deriving an upper bound on theRademacher Complexitythatdepends ontwokeyquantities: (i)theintrinsic dimension, which is a measure of isotropy, and (ii) the largest eigenvalue of the second moment (covariance) matrix ofthe distribution.

Under this perspective, we propose a token-based generator (i.e.,TokenGAN).