AI Paper Atlas

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

Evidential Knowledge Distillation

Liangyu Xiang, Junyu Gao, Changsheng Xu

ICCV 2025 · Label-efficient and robust learning › Knowledge distillation · approach: Distillation with causal and adaptive guidance · 8 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
replace vanilla KD loss with first-order Dirichlet-expectation loss L1st alongside L2nd (ablation)EKD (ResNet32x4->ResNet8x4) · CIFAR-100 val (ablation)top-1 accuracy: 76.34 -> 77.21Improved
replace MSE magnitude alignment with second-order Dirichlet KL L2nd alongside L1st (ablation)EKD (ResNet32x4->ResNet8x4) · CIFAR-100 val (ablation)top-1 accuracy: 76.83 -> 77.21Improved
replace MSE with L2nd when paired with vanilla KD (ablation)KD (ResNet32x4->ResNet8x4) · CIFAR-100 val (ablation)top-1 accuracy: 76.13 -> 76.34Improved
replace vanilla KD with L1st when paired with MSE (ablation)KD+MSE (ResNet32x4->ResNet8x4) · CIFAR-100 val (ablation)top-1 accuracy: 76.13 -> 76.83Improved
train teacher and student with evidential loss (EDL) instead of softmax cross-entropyResNet8x4 student · CIFAR-100 valtop-1 accuracy: 72.50 -> 72.77No clear effect
train MobileNetV2 student with EDL instead of softmaxMN-V2 student · CIFAR-100 valtop-1 accuracy: 64.60 -> 65.39Improved
replace vanilla KD with EKD (ResNet32x4->ResNet8x4)ResNet8x4 student · CIFAR-100 valtop-1 accuracy: 73.33 -> 77.21Improved
replace vanilla KD with EKD on ResNet50->MobileNetV1MobileNetV1 student · ImageNet valtop-1 accuracy: 70.50 -> 72.54Improved
replace vanilla KD with EKD on ResNet34->ResNet18ResNet18 student · ImageNet valtop-1 accuracy: 71.03 -> 71.73Improved
add EKD to RKD feature distillation (VGG13->VGG8)RKD · CIFAR-100 valtop-1 accuracy: 71.48 -> 74.99Improved
add EKD to RKD feature distillation (ResNet32x4->SHN-V2)RKD · CIFAR-100 valtop-1 accuracy: 73.21 -> 78.08Improved
add EKD to CAT-KD (VGG13->VGG8)CAT-KD · CIFAR-100 valtop-1 accuracy: 74.65 -> 74.91Improved

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