Permutation-Equivariant Flow Matching for Alignment-Free Neural Weight Generation
This work parameterize a flow-matching velocity field with a permutation-equivariant Graph Meta Network, enabling direct learning from independently trained networks without alignment, and shows how permutation equivariance enables learning from diverse collections of independently trained networks without permutation...
Arkadi Piven, Yam Eitan, Guy Bar-Shalom et al.
· 0 citations