AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

PADCLIP: Pseudo-labeling with Adaptive Debiasing in CLIP for Unsupervised Domain Adaptation

Zhengfeng Lai, Noranart Vesdapunt, Ning Zhou, Jun Wu, Cong Phuoc Huynh, Xuelu Li, Kah Kuen Fu, Chen-Nee Chuah

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment · 71 citations (OpenAlex)

Read the paper

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
fine-tune CLIP naively on source domain (ablation)CLIP ResNet-101 · VisDA-2017 target valaccuracy (%): 82.3% -> 55.6%Hurt
add Catastrophic Forgetting Measurement (CFM) to adjust learning rate (ablation)CLIP ResNet-101 fine-tuning · VisDA-2017 target valaccuracy (%): 55.6% -> 83.9%Improved
add pseudo-labeling on target domain (ablation)CLIP + CFM · VisDA-2017 target valaccuracy (%): 83.9% -> 86.0%Improved
add adaptive debiasing (CFM and DCM) to pseudo-labeling (ablation)CLIP + CFM + PL · VisDA-2017 target valaccuracy (%): 86.0% -> 88.5%Improved
add CFM to DebiasPL supervised loss (ablation)DebiasPL · VisDA-2017 target valaccuracy (%): 64.4% -> 85.9%Improved
add domain name to prompt (ablation)DebiasPL + CFM · VisDA-2017 target valaccuracy (%): 85.9% -> 86.5%Improved
add Dual Consistency Momentum to pseudo-label debiasing (ablation)DebiasPL + CFM · VisDA-2017 target valaccuracy (%): 87.8% -> 88.5%Improved
reduce pretraining data from 400M to 3M pairs (ablation)CLIP · VisDA-2017 target valaccuracy (%): 82.3% -> 59.1%Hurt
lower learning rate 50x or early stop instead of CFM (ablation)CLIP ResNet-101 · VisDA-2017 target valaccuracy (%): 55.6% -> 75.8%Improved
apply method to ResNet-101 with ImageNet-1K pretrainingPADCLIP · VisDA-2017accuracyNo clear effect
apply CFM to incremental learning and domain generalizationCLIP fine-tuning · incremental learning, domain generalizationaccuracy: +5.3%, +2.4%Improved

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