Fisher information geometry supplies parameterization-invariant metrics for MoE specialization dynamics and early failure prediction with strong empirical correlations.
Towards understanding the mixture-of-experts layer in deep learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Geometric Metrics for MoE Specialization: From Fisher Information to Early Failure Detection
Fisher information geometry supplies parameterization-invariant metrics for MoE specialization dynamics and early failure prediction with strong empirical correlations.