AI Paper Atlas

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

Simple Unsupervised Knowledge Distillation With Space Similarity

Aditya Singh, Haohan Wang

ECCV 2024 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 1 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 space similarity to cosine similarity loss (ResNet-18 student) (ablation)CoSS · ImageNet-1K linear evaltop-1 accuracy: 61.85 -> 62.35Improved
add cosine similarity to space similarity loss (EfficientNet-B0 student) (ablation)CoSS · ImageNet-1K linear evaltop-1 accuracy: 66.76 -> 67.36Improved
add k=15 nearest-neighbour sampling to batches (ablation)CoSS · ImageNet-1K linear eval (R18)top-1 accuracy: 61.89 -> 62.35Improved
weigh space similarity with lambda 0.5 or 1.0 rather than 0 (ablation)CoSS · ImageNet-1K linear eval (R18)top-1 accuracy: 61.85 -> 62.35Improved
lambda 0.5 vs 1.0 (ablation)CoSS · ImageNet-1K linear eval (R18)top-1 accuracy: 62.41 -> 62.35No clear effect
use CoSS instead of DINO-style soft-label distillation (ablation)unsupervised distillation · ImageNet-1K linear eval (R18)top-1 accuracy: 59.88 -> 62.35Improved
use space similarity instead of BatchNorm-based embedding matching (ablation)CIFAR-100 distillation · CIFAR-100 ResNet8x4/32x4top-1 accuracy: 72.22 -> 73.90Improved
add space similarity to SEED objective (ablation)SEED · ImageNet-1K linear evaltop-1 accuracy: 57.60 -> 58.30Improved
lambda choice on CIFAR-100 ResNet20/56 (ablation)CoSS · CIFAR-100top-1 accuracy: ResNet8x4: 72.05 -> 73.91; ResNet20/56: 70.58 -> 71.34 (best at 0.25), 70.68 at 1Mixed
CoSS vs PCD on dense predictionCoSS · MS-COCO and VOC 07+12AP: COCO AP bbox 32.2 -> 33.9; VOC AP50 79.4 -> 78.7Mixed
CoSS vs DisCo on STL-10 transferCoSS · STL-10 transfer (ResNet-18)top-1 accuracy: 95.04 -> 94.11Hurt
CoSS vs BINGO on ImageNet KNN-10 (ResNet-18)CoSS · ImageNet-1K KNN-10KNN-10 accuracy: 54.16 -> 53.78Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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