Transformer Fusion with Optimal Transport

arXiv:2310.05719v3 Announce Type: replace
Abstract: Fusion is a technique for merging multiple independently-trained neural networks in order to combine their capabilities. Past attempts have been restricted to the case of fully-connected, convolutional, and residual networks. This paper presents a systematic approach for fusing two or more transformer-based networks exploiting Optimal Transport to (soft-)align the various architectural components. We flesh out an abstraction for layer alignment, that can generalize to arbitrary architectures - in principle - and we apply this to …

arxiv cs.lg fusion stat.ml transformer transport type