AI Paper Atlas

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

Markov Knowledge Distillation: Make Nasty Teachers trained by Self-undermining Knowledge Distillation Fully Distillable

En-hui Yang, Linfeng Ye

ECCV 2024 · Label-efficient and robust learning › Knowledge distillation · approach: Distillation with causal and adaptive guidance · 4 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 Markov transform (co-learned) to distillation from a nasty teacher (ablation)MKD-KD · CIFAR-100, nasty R18 teacher, MobileNetV2 studenttop-1 accuracy (%): 42.76 -> 71.41Improved
add power transform on top of Markov transform (ablation)MKD-KD · CIFAR-100, nasty R18 teacher, MobileNetV2 studenttop-1 accuracy (%): 71.37 -> 71.41No clear effect
use MKD-KD instead of standard KD to distill a nasty teacherKD · CIFAR-100, nasty R18 teacher, MobileNetV2 studenttop-1 accuracy (%): 6.53 -> 71.41Improved
use MKD-KD instead of Skep., HTC, Avg. for nasty teachersnasty teacher distillation · CIFAR-10/100, TinyImageNettop-1 accuracy (%): MKD-KD above LS student in all pairs, up to 3.57% gainImproved
apply Markov transform to normal-teacher distillation (MKD-KD vs KD)KD · CIFAR-100 and TinyImageNet, normal teacherstop-1 accuracy (%): TinyImageNet R18/SV2: 58.09 -> 60.96; CIFAR-100 R18/MV2: 72.41 -> 72.31Mixed
apply Markov transform on top of DKD for normal teachersDKD · CIFAR-100 normal teacherstop-1 accuracy (%): 72.60 -> 73.04 (R18/MV2)Improved
use MKD-KD for cross-domain distillation instead of KDKD · Cifar-50-50 (nasty R18, MV2)top-1 accuracy (%): 4.04 -> 79.53Improved
existing logit-based KD methods applied to nasty teacherKD, DKD, DIST, NKD, MLD, Skep., HTC, Avg. · CIFAR-10, nasty R18 teachertop-1 accuracy (%): RC20: LS 92.53, none of the methods exceeds itHurt

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