# Generative Pseudo-label Refinement for Unsupervised Domain Adaptation

arXiv: https://arxiv.org/abs/2001.02950 · HTML: https://mlsanity.com/paper/2001.02950/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jan 2020.

## Built on by (most cited first, 24 of 24)

- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2109.06165/) (2021)
- [Balancing Discriminability and Transferability for Source-Free Domain Adaptation](https://mlsanity.com/paper/2206.08009/) (2022)
- [Dynamic Weighted Learning for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2103.13814/) (2021)
- [Source-Free Adaptation to Measurement Shift via Bottom-Up Feature Restoration](https://mlsanity.com/paper/2107.05446/) (2021)
- [Estimating Egocentric 3D Human Pose in the Wild with External Weak Supervision](https://mlsanity.com/paper/2201.07929/) (2022)
- [Unsupervised Domain Adaptation of Black-Box Source Models](https://mlsanity.com/paper/2101.02839/) (2021)
- [TimeMatch: Unsupervised Cross-Region Adaptation by Temporal Shift Estimation](https://mlsanity.com/paper/2111.02682/) (2021)
- [CAFA: Class-Aware Feature Alignment for Test-Time Adaptation](https://mlsanity.com/paper/2206.00205/) (2022)
- [Divide to Adapt: Mitigating Confirmation Bias for Domain Adaptation of Black-Box Predictors](https://mlsanity.com/paper/2205.14467/) (2022)
- [Enhancing Source-Free Domain Adaptive Object Detection with Low-confidence Pseudo Label Distillation](https://mlsanity.com/paper/2407.13524/) (2024)
- [In Search for a Generalizable Method for Source Free Domain Adaptation](https://mlsanity.com/paper/2302.06658/) (2023)
- [Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation](https://mlsanity.com/paper/2210.10019/) (2022)
- [Towards Accurate Cross-Domain In-Bed Human Pose Estimation](https://mlsanity.com/paper/2110.03578/) (2021)
- [DATA: Multi-Disentanglement based Contrastive Learning for Open-World Semi-Supervised Deepfake Attribution](https://mlsanity.com/paper/2505.04384/) (2025)
- [Adaptive Pseudo-Label Refinement by Negative Ensemble Learning for Source-Free Unsupervised Domain Adaptation](https://mlsanity.com/paper/2103.15973/) (2021)
- [Continual Source-Free Unsupervised Domain Adaptation](https://mlsanity.com/paper/2304.07374/) (2023)
- [Fusing Conditional Submodular GAN and Programmatic Weak Supervision](https://mlsanity.com/paper/2312.10366/) (2023)
- [Domain Adaptation Using Pseudo Labels](https://mlsanity.com/paper/2402.06809/) (2024)
- [Unsupervised domain adaptation via double classifiers based on high confidence pseudo label](https://mlsanity.com/paper/2105.04729/) (2021)
- [Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation](https://mlsanity.com/paper/2404.10574/) (2024)
- [Conditioning GAN Without Training Dataset](https://mlsanity.com/paper/2405.20687/) (2024)
- [Domain Adaptation via Feature Refinement](https://mlsanity.com/paper/2508.16124/) (2025)
- [Radar-APLANC: Unsupervised Radar-based Heartbeat Sensing via Augmented Pseudo-Label and Noise Contrast](https://mlsanity.com/paper/2511.08071/) (2025)

## Abstract (arXiv)

We investigate and characterize the inherent resilience of conditional Generative Adversarial Networks (cGANs) against noise in their conditioning labels, and exploit this fact in the context of Unsupervised Domain Adaptation (UDA). In UDA, a classifier trained on the labelled source set can be used to infer pseudo-labels on the unlabelled target set. However, this will result in a significant amount of misclassified examples (due to the well-known domain shift issue), which can be interpreted as noise injection in the ground-truth labels for the target set. We show that cGANs are, to some extent, robust against such "shift noise". Indeed, cGANs trained with noisy pseudo-labels, are able to filter such noise and generate cleaner target samples. We exploit this finding in an iterative procedure where a generative model and a classifier are jointly trained: in turn, the generator allows to sample cleaner data from the target distribution, and the classifier allows to associate better labels to target samples, progressively refining target pseudo-labels. Results on common benchmarks show that our method performs better or comparably with the unsupervised domain adaptation state of the art.