AI Paper Atlas

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

Class-Specific Semantic Reconstruction for Open Set Recognition

Hongzhi Huang, Yu Wang, Qinghua Hu, Ming‐Ming Cheng

TPAMI 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 111 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
use pixelwise softmax then average pooling instead of average pooling then softmax (ablation)CSSR · CIFAR10 unknown detection averageAUROC: 94.7 -> 97.4Improved
use MAE instead of MSE as reconstruction error (ablation)CSSR · CIFAR10 unknown detection averageAUROC: 96.9 -> 97.4Improved
replace linear classification layer with CSSR class-specific autoencoders (ablation)linear baseline · CIFAR10 unknown detection averageAUROC: 94.4 -> 97.4Improved
add pixelwise softmax strategy to linear classifier (ablation)Linear · CIFAR10 unknown detection averageAUROC: 94.4 -> 94.7Improved
replace CSSR with RCSSR on CIFAR10 unknown detection (ablation)CSSR · CIFAR10 unknown detection averageAUROC: 97.4 -> 97.3No clear effect
integrate RRE and feature magnitude into score for CSSR vs max softmax-like s1 (ablation)CSSR · ImageNet-30AUROC: 92.2 -> 94.8Improved
use fused score s_all instead of best individual score (ablation)CSSR · ImageNet-30AUROC: 95.5 -> 95.3No clear effect
add training-time CSSR layer vs baseline on CIFAR10 closed setbaseline · CIFAR10 closed setaccuracy: 95.1 -> 95.3Improved
replace CSSR with RCSSR on closed setCSSR · CIFAR10 closed setaccuracy: 95.3 -> 95.6Improved
use CSSR instead of plain baseline on ImageNet-1000 splitbaseline · ImageNet-1000 splitTNR@TPR95: 51.8 -> 62.4Improved
use RCSSR instead of CSSR on large-scale iNaturalist unknownsCSSR · iNaturalist unknownAUROC: 93.8 -> 94.5Improved
use CSSR instead of ARPL on CIFAR+10ARPL · CIFAR+10 unknown detectionAUROC: 97.1 -> 96.3Hurt

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