AI Paper Atlas

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

Two-Way Multi-Label Loss

Takumi Kobayashi

CVPR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 43 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
two-way loss (sample-wise plus class-wise) instead of single-direction losses (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 74.11 vs 73.06 (class-only) and 67.18 (sample-only)Improved
positive-side temperature T_P=4 in the log-sum-exp margin loss (vs T_P=1) (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 74.11 vs 72.30Improved
setting hard positive min via T_P→0 (no log-sum-exp smoothing) (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 71.03 vs 74.11 bestHurt
logit bias (multi-similarity style) instead of temperature margin (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 73.44 vs 74.11 with temperatureHurt
replace softmax-style loss with proposed loss in two-way form (softmax two-way comparison) (ablation)softmax loss in two ways · MSCOCOmAP@class: 69.19 vs 74.11Hurt
joint-way loss over all class and sample entries instead of two-way (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 71.82 vs 74.11Hurt
multi-similarity loss instead of proposed loss (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@class: 71.27 vs 74.11Hurt
bucket size M reduced from 128 to 16 (ablation)class-wise loss in two-way loss · MSCOCOmAP@class: 70.76 vs 74.11Hurt
proposed two-way loss vs BCE/Focal/ASL/softmax on multi-label benchmarks (ResNet-50, ResNeXt50, DenseNet169, RegNetY32gf)multi-label classification · MSCOCO, VISPR, VAW, WIDER, VOC2007, VOC2012mAP@class and mAP@sample: e.g. MSCOCO mAP@class 74.11 vs ASL 70.92 (ResNet-50)Improved
ASL instead of proposed loss on ImageNet-transferResNet-50 ImageNet-trained features · Transfer (Aircraft, 10 datasets)accuracy: Aircraft 27.42 vs 44.01 oursHurt
two-way loss vs softmax on single-label ImageNet trainingResNet-50 ImageNet-pretrained features · ImageNet top-1 and transfertop-1 accuracy: 76.29 vs 76.32 top-1; transfer gains on Aircraft (44.01 vs 39.18)No clear effect
logit bias on the multi-similarity-style loss vs no bias (ablation)multi-label loss on ResNet-50 · MSCOCOmAP@sample: 86.19-86.25 across eps 0.5-2 vs 86.66 oursNo clear effect

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