AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

MoE-KD: Your Teacher Model is Worth Mixture-of-Experts for Knowledge Distillation

Kuo Shi, Wenjie Zhu, Bo Peng

ECCV 2026 · Label-efficient and robust learning › Knowledge distillation · approach: Distillation with causal and adaptive guidance · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add teacher-classifier gating instead of uniform gating (ablation)MoE-KD · CIFAR-100 WRN-40-2 -> WRN-40-1top-1 accuracy (%): 73.87 -> 75.21Improved
add teacher-classifier gating instead of uniform gating (ablation)MoE-KD · CIFAR-100 ResNet50 -> MobileNetV2top-1 accuracy (%): 69.28 -> 71.38Improved
reuse pre-trained teacher classifier as gating instead of learning it from scratch (ablation)MoE-KD · CIFAR-100 WRN-40-2 -> WRN-40-1top-1 accuracy (%): 74.43 -> 75.21Improved
reuse pre-trained teacher classifier as gating instead of learning it from scratch (ablation)MoE-KD · CIFAR-100 ResNet50 -> MobileNetV2top-1 accuracy (%): 70.62 -> 71.38Improved
derive expert embeddings from teacher features instead of learning from scratch (ablation)MoE-KD · CIFAR-100 WRN-40-2 -> WRN-40-1top-1 accuracy (%): 74.66 -> 75.21Improved
derive expert embeddings from teacher features instead of learning from scratch (ablation)MoE-KD · CIFAR-100 ResNet50 -> MobileNetV2top-1 accuracy (%): 70.65 -> 71.38Improved
use soft aggregation instead of hard aggregation (ablation)MoE-KD · CIFAR-100 WRN-40-2 -> WRN-40-1top-1 accuracy (%): 74.95 -> 75.21Improved
use soft aggregation instead of hard aggregation (ablation)MoE-KD · CIFAR-100 ResNet50 -> MobileNetV2top-1 accuracy (%): 71.16 -> 71.38Improved
use Bayes-optimal EM E-step instead of direct teacher assignment (ablation)MoE-KD · CIFAR-100 WRN-40-2 -> WRN-40-1top-1 accuracy (%): 74.79 -> 75.21Improved
use Bayes-optimal EM E-step instead of direct teacher assignment (ablation)MoE-KD · CIFAR-100 ResNet50 -> MobileNetV2top-1 accuracy (%): 70.94 -> 71.38Improved
use MoE-KD instead of DiffKDMoE-KD · ImageNet-1K ResNet34 -> ResNet18top-1 accuracy (%): 72.22 -> 73.15Improved
use MoE-KD instead of DiffKDMoE-KD · ImageNet-1K ResNet50 -> MobileNetV1top-1 accuracy (%): 73.62 -> 73.41Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy