AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Echoes of the Past: Boosting Long-tail Recognition via Reflective Learning

Qihao Zhao, Yalun Dai, Shen Lin, Wei Hu, Fan Zhang, Jun Liu

ECCV 2024 · Label-efficient and robust learning › Recognition and long-tail · approach: Decoupled and calibrated long-tail training

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 Knowledge Review (KR) consistency distillation (ablation)RIDE · ImageNet-LTaccuracy: 56.3 -> 58.0Improved
add Knowledge Summary (KS) soft class-correlation labels (ablation)RIDE · ImageNet-LTaccuracy: 56.3 -> 58.4Improved
add Knowledge Correction (gradient projection) on top of KR+KS (ablation)RIDE · ImageNet-LTaccuracy: 58.6 -> 59.0Improved
add KC on top of KR+KS (ablation)SADE · iNaturalist 2018accuracy: 73.8 -> 74.2Improved
add Correctly Classified Instances (CCI) filtering to KR (ablation)BSCE+RL · CIFAR-100-LT IF=100top-1 accuracyImproved
replace KL-based KR with direct MSE logit matching (ablation)BSCE+RL · CIFAR-100-LT IF=100top-1 accuracyHurt
set temperature tau to 2 (ablation)RL · CIFAR-100-LT and ImageNet-LTaccuracyImproved
add RL to SADESADE · ImageNet-LT few-shotfew-shot accuracy: 43.5 -> 47.8Improved
add RL to Softmax baselineSoftmax · CIFAR-100-LT IF=100top-1 accuracy: 41.4 -> 41.9Improved
add RL to RIDE across backbonesRIDE · ImageNet-LToverall accuracy: 56.3 -> 59.1 (Swin-T)Improved
KR alone slightly reduces head-class accuracy (ablation)RL · ImageNet-LTmany-shot accuracyMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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