Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.21521 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908470884696064 |
|---|---|
| author | Narayanan, Athmanarayanan Lakshmi Machireddy, Amrutha Krishnan, Ranganath |
| author_facet | Narayanan, Athmanarayanan Lakshmi Machireddy, Amrutha Krishnan, Ranganath |
| contents | Active Learning (AL) has emerged as a powerful approach for minimizing labeling costs by selectively sampling the most informative data for neural network model development. Effective AL for large-scale vision-language models necessitates addressing challenges in uncertainty estimation and efficient sampling given the vast number of parameters involved. In this work, we introduce a novel parameter-efficient learning methodology that incorporates uncertainty calibration loss within the AL framework. We propose a differentiable loss function that promotes uncertainty calibration for effectively selecting fewer and most informative data samples for fine-tuning. Through extensive experiments across several datasets and vision backbones, we demonstrate that our solution can match and exceed the performance of complex feature-based sampling techniques while being computationally very efficient. Additionally, we investigate the efficacy of Prompt learning versus Low-rank adaptation (LoRA) in sample selection, providing a detailed comparative analysis of these methods in the context of efficient AL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_21521 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Optimizing Active Learning in Vision-Language Models via Parameter-Efficient Uncertainty Calibration Narayanan, Athmanarayanan Lakshmi Machireddy, Amrutha Krishnan, Ranganath Computer Vision and Pattern Recognition Active Learning (AL) has emerged as a powerful approach for minimizing labeling costs by selectively sampling the most informative data for neural network model development. Effective AL for large-scale vision-language models necessitates addressing challenges in uncertainty estimation and efficient sampling given the vast number of parameters involved. In this work, we introduce a novel parameter-efficient learning methodology that incorporates uncertainty calibration loss within the AL framework. We propose a differentiable loss function that promotes uncertainty calibration for effectively selecting fewer and most informative data samples for fine-tuning. Through extensive experiments across several datasets and vision backbones, we demonstrate that our solution can match and exceed the performance of complex feature-based sampling techniques while being computationally very efficient. Additionally, we investigate the efficacy of Prompt learning versus Low-rank adaptation (LoRA) in sample selection, providing a detailed comparative analysis of these methods in the context of efficient AL. |
| title | Optimizing Active Learning in Vision-Language Models via Parameter-Efficient Uncertainty Calibration |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.21521 |