# F^2TTA: Free-Form Test-Time Adaptation on Cross-Domain Medical Image Classification via Image-Level Disentangled Prompt Tuning

arXiv: https://arxiv.org/abs/2507.02437 · HTML: https://mlsanity.com/paper/2507.02437/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 39 references that are in the map.
- First posted on arXiv 3 Jul 2025.

## Builds on (most cited first, 17 of 17)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Transformers in Medical Imaging: A Survey](https://mlsanity.com/paper/2201.09873/) (2022)
- [MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation](https://mlsanity.com/paper/2212.01322/) (2022)
- [MS-Net: Multi-Site Network for Improving Prostate Segmentation with Heterogeneous MRI Data](https://mlsanity.com/paper/2002.03366/) (2020)
- [Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptation](https://mlsanity.com/paper/2212.04145/) (2022)
- [Visual Prompt Tuning for Test-time Domain Adaptation](https://mlsanity.com/paper/2210.04831/) (2022)
- [Unsupervised Domain Adaptation via Disentangled Representations: Application to Cross-Modality Liver Segmentation](https://mlsanity.com/paper/1907.13590/) (2019)
- [Exploring Sparse Visual Prompt for Domain Adaptive Dense Prediction](https://mlsanity.com/paper/2303.09792/) (2023)
- [Contrastive Semi-supervised Learning for Domain Adaptive Segmentation Across Similar Anatomical Structures](https://mlsanity.com/paper/2208.08605/) (2022)
- [CDDSA: Contrastive Domain Disentanglement and Style Augmentation for Generalizable Medical Image Segmentation](https://mlsanity.com/paper/2211.12081/) (2022)
- [Comprehensive Generative Replay for Task-Incremental Segmentation with Concurrent Appearance and Semantic Forgetting](https://mlsanity.com/paper/2406.19796/) (2024)
- [FunOTTA: On-the-Fly Adaptation on Cross-Domain Fundus Image via Stable Test-time Training](https://mlsanity.com/paper/2407.04396/) (2024)

## Abstract (arXiv)

Test-Time Adaptation (TTA) has emerged as a promising solution for adapting a source model to unseen medical sites using unlabeled test data, due to the high cost of data annotation. Existing TTA methods consider scenarios where data from one or multiple domains arrives in complete domain units. However, in clinical practice, data usually arrives in domain fragments of arbitrary lengths and in random arrival orders, due to resource constraints and patient variability. This paper investigates a practical Free-Form Test-Time Adaptation (F2TTA) task, where a source model is adapted to such free-form domain fragments, with shifts occurring between fragments unpredictably. In this setting, these shifts could distort the adaptation process. To address this problem, we propose a novel Image-level Disentangled Prompt Tuning (I-DiPT) framework. I-DiPT employs an image-invariant prompt to explore domain-invariant representations for mitigating the unpredictable shifts, and an image-specific prompt to adapt the source model to each test image from the incoming fragments. The prompts may suffer from insufficient knowledge representation since only one image is available for training. To overcome this limitation, we first introduce Uncertainty-oriented Masking (UoM), which encourages the prompts to extract sufficient information from the incoming image via masked consistency learning driven by the uncertainty of the source model representations. Then, we further propose a Parallel Graph Distillation (PGD) method that reuses knowledge from historical image-specific and image-invariant prompts through parallel graph networks. Experiments on breast cancer and glaucoma classification demonstrate the superiority of our method over existing TTA approaches in F2TTA. Code is available at https://github.com/mar-cry/F2TTA.