AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Decouple Then Converge: Handling Unknown Unlabeled Distributions in Long-Tailed Semi-Supervised Learning

Kai Gan, Tong Wei, Min-Ling Zhang

TPAMI 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 2 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 alignment of the standard branch to the balanced-branch distribution (ablation)DTC · CIFAR-10-LT reversedaccuracy: 66.8 -> 86.6Improved
add balanced pseudo-labels (logit-adjusted standard-branch outputs) (ablation)DTC · CIFAR-10-LT consistentaccuracy: 82.2 -> 84.1Improved
add balanced softmax in the balanced branch (ablation)DTC · CIFAR-100-LT consistentaccuracy: 51.1 -> 52.0Improved
add importance weighting of pseudo-labels (ablation)DTC · CIFAR-100-LT uniformaccuracy: 58.3 -> 60.4Improved
add post-hoc logit adjustment (tau3=2) (ablation)DTC · CIFAR-10-LT consistentaccuracy: 77.8 -> 84.1Improved
add post-hoc adjustment (tau3=2) to ACR (ablation)ACR · CIFAR-10-LT consistentaccuracy: 81.6 -> 82.0Improved
add cross-branch pseudo-labeling (CBPL) to ACR, before post-hoc (ablation)ACR · CIFAR-10-LT consistentaccuracy: 81.6 -> 78.1Hurt
add CBPL to ACR, before post-hoc (ablation)ACR · CIFAR-10-LT uniformaccuracy: 92.1 -> 92.8Improved
add decoupled distribution alignment on top of CBPL (ablation)DTC · CIFAR-10-LT uniformaccuracy: 92.8 -> 93.5Improved
add importance weighting on top of CBPL and DDA (ablation)DTC · CIFAR-10-LT uniformaccuracy: 93.5 -> 93.8Improved
use DTC instead of ACR on random unlabeled distributionsFixMatch · CIFAR-100-LT random distributionsaverage accuracy: 43.6 -> 45.7Improved
add DTC to FixMatch (dual branch)FixMatch · CIFAR-100-LTaccuracy and training time: 49.7 -> 57.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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