AI Paper Atlas

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

Revisiting Knowledge Distillation: An Inheritance and Exploration Framework

Zhen Huang, Xu Shen, Jun Xing, Tongliang Liu, Xinmei Tian, Houqiang Li, Bing Deng, Jianqiang Huang, Xian-Sheng Hua

CVPR 2021 · Label-efficient and robust learning › Knowledge distillation · approach: Distillation with causal and adaptive guidance · 29 citations (OpenAlex)

Check 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
Apply IE-KD (IE-OD) when transferring from a smaller teacher to a larger studentStudent networks · CIFAR10, four small-teacher-to-large-student architecture pairsTop-1 test error (%): 5.44; 5.68; 4.24; 5.35Improved
Apply IE-KD (IE-OD) when transferring from a smaller teacher to a larger studentResNet-34 · ImageNet, ResNet-18 teacher to ResNet-34 studentTop-1 test error (%): 26.73 -> 25.54Improved
Apply IE-KD (IE-OD) when transferring from a smaller teacher to a larger studentResNet-34 · ImageNet, ResNet-18 teacher to ResNet-34 studentTop-5 test error (%): 8.57 -> 8.06Improved
Use IE-OD instead of the strongest listed prior distillation methodStudent networks · CIFAR100, four teacher-to-student architecture pairsTop-1 test error (%): 6.73 -> 6.33; 6.77 -> 6.39; 4.62 -> 4.45; 7.48 -> 7.16Improved

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