AI Paper Atlas

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

Graph-Driven Domain Co-Adaptation for Cross-Domain Image Quality Assessment

Shun Zhu, Xichen Yang, Yan Zhang, Tianshu Wang, Zhongyuan Mao, Tianyin Li, Zhuoyan Sun, Xiaobo Shen

AAAI 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 1 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 GKS, GLA and CAL together (ablation)GDCIQA baseline · LIVEC (train KADID)PLCC: 0.576 -> 0.756Improved
add CAL co-adaptation learning to GKS+GLA (ablation)GDCIQA (GKS+GLA) · LIVEC (train KADID)PLCC: 0.624 -> 0.756Improved
add GKS graph knowledge sharing to GLA+CAL (ablation)GDCIQA (GLA+CAL) · AGIQA-3k (train KADID)PLCC: 0.721 -> 0.794Improved
add GLA to GKS+CAL (ablation)GDCIQA (GKS+CAL) · LIVEC (train KADID)PLCC: 0.730 -> 0.756Improved
add GKS alone (ablation)no-module baseline · LIVEC (train KADID)PLCC: 0.576 -> 0.607Improved
add 1DCNN in GLA (ablation)GLA without cross-attention · LIVEC (train KADID)PLCC: 0.730 -> 0.741Improved
add cross-attention to GLA with 1DCNN (ablation)GLA · LIVEC (train KADID)PLCC: 0.741 -> 0.756Improved
GDCIQA vs DGQADGQA · LIVEC (train KADID)SRCC: 0.696 -> 0.753Improved
GDCIQA vs StyleAMStyleAM · BID (train KADID)SRCC: 0.637 -> 0.787Improved
use joint loss L1+L2 instead of only L1 or L2 (ablation)CAL · LIVEC convergenceepochs to converge: 27 -> 17 epochsImproved

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