Empowering Source-Free Domain Adaptation via MLLM-Guided Reliability-Based Curriculum Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914234417283072 |
|---|---|
| author | Chen, Dongjie Patwari, Kartik Lai, Zhengfeng Zhu, Xiaoguang Cheung, Sen-ching Chuah, Chen-Nee |
| author_facet | Chen, Dongjie Patwari, Kartik Lai, Zhengfeng Zhu, Xiaoguang Cheung, Sen-ching Chuah, Chen-Nee |
| contents | Existing SFDA methods struggle to fully use pre-trained knowledge and often rely on a single model's predictions or handcrafted prompts, limiting robustness under domain shift. Multimodal Large Language Models (MLLMs) offer a promising alternative: they encode rich visual-semantic knowledge and generalize well without task-specific tuning. However, their use in SFDA is hindered by instruction-following failures, inconsistent outputs, and high inference costs. We propose Reliability-based Curriculum Learning (RCL), a novel framework that distills robust supervision from multiple frozen MLLMs into a compact target model. RCL organizes adaptation as a three-stage curriculum that progressively incorporates pseudo-labels based on inter-model agreement and model confidence, enabling stable and noise-aware training. Our approach achieves state-of-the-art performance on standard SFDA datasets, Office-Home, DomainNet-126, and VisDA-C, outperforming zero-shot MLLMs, their ensembles, all without accessing source data or tuning foundation models. Our code is available at: https://github.com/Dong-Jie-Chen/RCL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_18376 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Empowering Source-Free Domain Adaptation via MLLM-Guided Reliability-Based Curriculum Learning Chen, Dongjie Patwari, Kartik Lai, Zhengfeng Zhu, Xiaoguang Cheung, Sen-ching Chuah, Chen-Nee Machine Learning Computer Vision and Pattern Recognition Existing SFDA methods struggle to fully use pre-trained knowledge and often rely on a single model's predictions or handcrafted prompts, limiting robustness under domain shift. Multimodal Large Language Models (MLLMs) offer a promising alternative: they encode rich visual-semantic knowledge and generalize well without task-specific tuning. However, their use in SFDA is hindered by instruction-following failures, inconsistent outputs, and high inference costs. We propose Reliability-based Curriculum Learning (RCL), a novel framework that distills robust supervision from multiple frozen MLLMs into a compact target model. RCL organizes adaptation as a three-stage curriculum that progressively incorporates pseudo-labels based on inter-model agreement and model confidence, enabling stable and noise-aware training. Our approach achieves state-of-the-art performance on standard SFDA datasets, Office-Home, DomainNet-126, and VisDA-C, outperforming zero-shot MLLMs, their ensembles, all without accessing source data or tuning foundation models. Our code is available at: https://github.com/Dong-Jie-Chen/RCL. |
| title | Empowering Source-Free Domain Adaptation via MLLM-Guided Reliability-Based Curriculum Learning |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.18376 |