Universal Domain Adaptation On Visda2017

Metrics

H-score

Source-free

Results

Performance results of various models on this benchmark

			Paper Title
TASC	90.36	no	Target Semantics Clustering via Text Representations for Robust Universal Domain Adaptation
LEAD	76.6	yes	LEAD: Learning Decomposition for Source-free Universal Domain Adaptation
GLC	73.1	yes	Upcycling Models under Domain and Category Shift
MLNet	69.9	no	MLNet: Mutual Learning Network with Neighborhood Invariance for Universal Domain Adaptation
UniAM	65.18	no	Universal Domain Adaptation via Compressive Attention Matching
SAN	60.1	no	Boosting Novel Category Discovery Over Domains with Soft Contrastive Learning and All in One Classifier
UMAD	58.3	yes	UMAD: Universal Model Adaptation under Domain and Category Shift
UniOT	57.32	no	Unified Optimal Transport Framework for Universal Domain Adaptation
GATE	56.4	no	Geometric Anchor Correspondence Mining With Uncertainty Modeling for Universal Domain Adaptation
OVANet	53.1	no	OVANet: One-vs-All Network for Universal Domain Adaptation
SHOT-O	44.0	yes	Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation
DCC	43.0	no	Domain Consensus Clustering for Universal Domain Adaptation
UAN	34.8	no	Universal Domain Adaptation

0 of 13 row(s) selected.

Universal Domain Adaptation On Visda2017 | SOTA | HyperAI