AI Paper Atlas

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

Knowledge Distillation with Refined Logits

Wujie Sun, Defang Chen, Siwei Lyu, Genlang Chen, Chun Chen, Can Wang

ICCV 2025 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 25 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 sample confidence distillation to CE (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 72.50 -> 73.55Improved
add masked correlation (M_ge) to CE (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 72.50 -> 75.64Improved
add masked correlation (M_g) to CE (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 72.50 -> 75.50Improved
add sample confidence to masked correlation with M_ge (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 75.64 -> 76.64Improved
add sample confidence to masked correlation with M_g (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 75.50 -> 75.53No clear effect
mask classes with M_ge instead of M_g in full RLD (ablation)RLD · CIFAR-100 ResNet32x4 -> ResNet8x4top-1 accuracy (%): 75.53 -> 76.64Improved
use RLD instead of DKD with inferior teacher ResNet56RLD · CIFAR-100 ResNet56 -> WRN-40-2top-1 accuracy (%): 77.34 -> 78.03Improved
use RLD instead of DKD with inferior teacher VGG13RLD · CIFAR-100 VGG13 -> WRN-40-2top-1 accuracy (%): 77.45 -> 77.88Improved
use RLD instead of DKD with LSKD logit standardizationRLD · CIFAR-100 ResNet50 -> MN-V2top-1 accuracy (%): 70.45 -> 71.06Improved
use RLD instead of DKDRLD · ImageNet ResNet50 -> MN-V1top-1 accuracy (%): 72.05 -> 72.75Improved
use RLD instead of DKDRLD · CIFAR-100 ResNet110 -> ResNet32top-1 accuracy (%): 74.11 -> 74.02Hurt
use RLD instead of DKDRLD · CIFAR-100 ResNet110 -> ResNet20top-1 accuracy (%): 71.06 -> 71.67Improved

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