AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910094838464512 |
|---|---|
| author | Deb, Prantik Dhondy, Srimanth Ramakrishna, N. Kapoor, Anu Bapi, Raju S. Chakraborti, Tapabrata |
| author_facet | Deb, Prantik Dhondy, Srimanth Ramakrishna, N. Kapoor, Anu Bapi, Raju S. Chakraborti, Tapabrata |
| contents | Chest X-ray (CXR) segmentation is an important step in computer-aided diagnosis, yet deploying large foundation models in clinical settings remains challenging due to computational constraints. We propose AdaLoRA-QAT, a two-stage fine-tuning framework that combines adaptive low-rank encoder adaptation with full quantization-aware training. Adaptive rank allocation improves parameter efficiency, while selective mixed-precision INT8 quantization preserves structural fidelity crucial for clinical reliability. Evaluated across large-scale CXR datasets, AdaLoRA-QAT achieves 95.6% Dice, matching full-precision SAM decoder fine-tuning while reducing trainable parameters by 16.6\times and yielding 2.24\times model compression. A Wilcoxon signed-rank test confirms that quantization does not significantly degrade segmentation accuracy. These results demonstrate that AdaLoRA-QAT effectively balances accuracy, efficiency, and structural trust-worthiness, enabling compact and deployable foundation models for medical image segmentation. Code and pretrained models are available at: https://prantik-pdeb.github.io/adaloraqat.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_01167 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation Deb, Prantik Dhondy, Srimanth Ramakrishna, N. Kapoor, Anu Bapi, Raju S. Chakraborti, Tapabrata Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Chest X-ray (CXR) segmentation is an important step in computer-aided diagnosis, yet deploying large foundation models in clinical settings remains challenging due to computational constraints. We propose AdaLoRA-QAT, a two-stage fine-tuning framework that combines adaptive low-rank encoder adaptation with full quantization-aware training. Adaptive rank allocation improves parameter efficiency, while selective mixed-precision INT8 quantization preserves structural fidelity crucial for clinical reliability. Evaluated across large-scale CXR datasets, AdaLoRA-QAT achieves 95.6% Dice, matching full-precision SAM decoder fine-tuning while reducing trainable parameters by 16.6\times and yielding 2.24\times model compression. A Wilcoxon signed-rank test confirms that quantization does not significantly degrade segmentation accuracy. These results demonstrate that AdaLoRA-QAT effectively balances accuracy, efficiency, and structural trust-worthiness, enabling compact and deployable foundation models for medical image segmentation. Code and pretrained models are available at: https://prantik-pdeb.github.io/adaloraqat.github.io/ |
| title | AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.01167 |