A router-norm and variance-based bit allocation strategy for quantizing MoE models that claims higher accuracy and lower cost than prior mixed-precision methods.
21 Published as a conference paper at ICLR 2026 Definitions: For anyq∈ P\{o 1, o2}, we define the activation of the experts∈[k]byqas, σ(s,t) q :=Pm r=1 ReLU(⟨w(s,t) r , q⟩)
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
-
Efficient Quantization of Mixture-of-Experts with Theoretical Generalization Guarantees
A router-norm and variance-based bit allocation strategy for quantizing MoE models that claims higher accuracy and lower cost than prior mixed-precision methods.