SSMoE uses eigenvectors of expert weights via SVD to build training-free non-collapsing routers for SMoE models in language and vision tasks.
Eigenvectors of Experts are Training-free Non-collapsing Routers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Eigenvectors of Experts are Training-free Non-collapsing Routers
SSMoE uses eigenvectors of expert weights via SVD to build training-free non-collapsing routers for SMoE models in language and vision tasks.