AI Paper Atlas

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

From Softmax to Dirichlet: Evidential Learning for Semi-supervised Semantic Segmentation

Huayu Mai, Rui Sun, Yujia Chen, Wangkai Li, Bingzhou Wang, Aibing Li, Zhangyu He, Yuan Wang

CVPR 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Consistency training for semi-supervised segmentation

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
replace softmax-score filtering with Dirichlet evidential uncertainty (ESS) (ablation)UniMatch · Pascal classic 1/16, ResNet-101mIoU: 75.2 -> 76.9Improved
add hyper-evidence (collective evidence) on top of ESS (ablation)UniMatch + ESS · Pascal classic 1/16, ResNet-101mIoU: 76.9 -> 77.5Improved
replace softmax-score filtering with Dirichlet evidential uncertainty (ESS) (ablation)UniMatch · Pascal classic 1/4, ResNet-101mIoU: 78.8 -> 79.5Improved
add hyper-evidence on top of ESS (ablation)UniMatch + ESS · Pascal classic 1/4, ResNet-101mIoU: 79.5 -> 80.0Improved
integrate HESS into UniMatchUniMatch · Pascal high-quality 1/16, ResNet-101mIoU: 75.2 -> 77.5Improved
integrate HESS into UniMatch V2UniMatch V2 · Pascal high-quality 1/16, DINOv2-SmIoU: 79.0 -> 80.9Improved
integrate HESS into UniMatchUniMatch · Cityscapes 1/16, ResNet-101mIoU: 76.7 -> 78.0Improved
integrate HESS into UniMatch V2UniMatch V2 · Cityscapes 1/16, DINOv2-SmIoU: 80.6 -> 81.8Improved
integrate HESS into UniMatchUniMatch · COCO 1/512, XC-65mIoU: 31.9 -> 33.8Improved
integrate HESS into UniMatch V2UniMatch V2 · COCO 1/512, DINOv2-SmIoU: 39.3 -> 41.4Improved
raise softmax threshold from 0.95 to 0.99 (ablation)UniMatch softmax filtering · Pascal classic 1/16, ResNet-101mIoU: 75.2 -> 73.9Hurt
set uncertainty threshold too low (0.10) or too high (0.40) instead of 0.25 (ablation)HESS uncertainty filtering · Pascal classic 1/16, ResNet-101mIoU: 77.5 -> 74.1 (0.10), 74.5 (0.40)Hurt

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