# Analysis of Pseudo-Labeling for Online Source-Free Universal Domain Adaptation

arXiv: https://arxiv.org/abs/2504.11992 · HTML: https://mlsanity.com/paper/2504.11992/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 16 references that are in the map.
- First posted on arXiv 16 Apr 2025.

## Builds on (most cited first, 12 of 12)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Universal Source-Free Domain Adaptation](https://mlsanity.com/paper/2004.04393/) (2020)
- [Towards Inheritable Models for Open-Set Domain Adaptation](https://mlsanity.com/paper/2004.04388/) (2020)
- [Upcycling Models under Domain and Category Shift](https://mlsanity.com/paper/2303.07110/) (2023)
- [LEAD: Learning Decomposition for Source-free Universal Domain Adaptation](https://mlsanity.com/paper/2403.03421/) (2024)
- [UMAD: Universal Model Adaptation under Domain and Category Shift](https://mlsanity.com/paper/2112.08553/) (2021)
- [Open-Set Hypothesis Transfer with Semantic Consistency](https://mlsanity.com/paper/2010.00292/) (2020)
- [GLC++: Source-Free Universal Domain Adaptation through Global-Local Clustering and Contrastive Affinity Learning](https://mlsanity.com/paper/2403.14410/) (2024)
- [Memory-Efficient Pseudo-Labeling for Online Source-Free Universal Domain Adaptation using a Gaussian Mixture Model](https://mlsanity.com/paper/2407.14208/) (2024)
- [COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation](https://mlsanity.com/paper/2308.10450/) (2023)
- [COMET: Contrastive Mean Teacher for Online Source-Free Universal Domain Adaptation](https://mlsanity.com/paper/2401.17728/) (2024)

## Abstract (arXiv)

A domain (distribution) shift between training and test data often hinders the real-world performance of deep neural networks, necessitating unsupervised domain adaptation (UDA) to bridge this gap. Online source-free UDA has emerged as a solution for practical scenarios where access to source data is restricted and target data is received as a continuous stream. However, the open-world nature of many real-world applications additionally introduces category shifts meaning that the source and target label spaces may differ. Online source-free universal domain adaptation (SF-UniDA) addresses this challenge. Existing methods mainly rely on self-training with pseudo-labels, yet the relationship between pseudo-labeling and adaptation outcomes has not been studied yet. To bridge this gap, we conduct a systematic analysis through controlled experiments with simulated pseudo-labeling, offering valuable insights into pseudo-labeling for online SF-UniDA. Our findings reveal a substantial gap between the current state-of-the-art and the upper bound of adaptation achieved with perfect pseudo-labeling. Moreover, we show that a contrastive loss enables effective adaptation even with moderate pseudo-label accuracy, while a cross-entropy (CE) loss, though less robust to pseudo-label errors, achieves superior results when pseudo-labeling approaches perfection. Lastly, our findings indicate that pseudo-label accuracy is in general more crucial than quantity, suggesting that prioritizing fewer but high-confidence pseudo-labels is beneficial. Overall, our study highlights the critical role of pseudo-labeling in (online) SF-UniDA and provides actionable insights to drive future advancements in the field. Our code is available at https://github.com/pascalschlachter/PLAnalysis.