Closing the Confusion Loop: CLIP-Guided Alignment for Source-Free Domain Adaptation
Shanshan Wang, Ziying Feng, Xiaozheng Shen, Xun Yang*, Pichao Wang, Zhenwei He, Xingyi Zhang · 2026
Read Closing the Confusion Loop: CLIP-Guided Alignment for Source-Free Domain Adaptation by Shanshan Wang, Ziying Feng, Xiaozheng Shen, Xun Yang*, Pichao Wang, Zhenwei He, Xingyi Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-08730.txt.
Description
Adapting a trained model to a new domain without the original data is valuable for privacy, but fine-grained classes confuse the process. Shanshan Wang and six co-authors study asymmetric, shifting class confusion, where similar classes are misclassified unequally over time. Their arXiv paper proposes CLIP-Guided Alignment to model and mitigate that confusion explicitly.
Three parts do the work. One detects directional confusion pairs from source-model predictions on the target. Another builds confusion-aware text prompts with CLIP, such as a truck that looks like a bus, for context-sensitive pseudo-labeling. A third aligns confusion-guided feature banks by contrastive learning. Tests show consistent wins over leading source-free methods, especially on confusion-prone fine-grained tasks. Readers get a clear case for modeling confusion rather than ignoring it.
Catalog record
| CREATOR | Shanshan Wang, Ziying Feng, Xiaozheng Shen, Xun Yang*, Pichao Wang, Zhenwei He, Xingyi Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-08730