Colla-Q: Toward Collaborative Experts in MoE Quantization via Minimax Precision Balancing
This work proposes Colla-Q, a bit-allocation framework to maintain balanced performance across experts through an activation-entropy-based bit-width allocation algorithm that encourages each expert to operate collaboratively in the quantized model, thereby improving the overall MoE performance and reducing the dependen...
Eunjun Shin, Jongbin Ryu
· 0 citations