Various adaptation methods, such as LoRA, prompts, and adapters, have been proposed to enhance the performance of pre-trained vision-language models in specific domains. The robustness of these adaptation methods against distribution shifts have not been studied. In this study, we assess the robustness of 11 widely-used adaptation methods across 4 vision-language datasets under multimodal corruptions. Concretely, we introduce 7 benchmark datasets, including 96 visual and 87 textual corruptions, to investigate the robustness of different adaptation methods, the impact of available adaptation examples, and the influence of trainable parameter size during adaptation. Our analysis reveals that: 1) Adaptation methods are more sensitive to text corruptions than visual corruptions. 2) Full fine-tuning does not consistently provide the highest robustness; instead, adapters can achieve better robustness with comparable clean performance. 3) Contrary to expectations, our findings indicate that increasing the number of adaptation data and parameters does not guarantee enhanced robustness; instead it results in even lower robustness. We hope this study could benefit future research in the development of robust multimodal adaptation methods. The benchmark, code, and dataset used in this study can be accessed at https://adarobustness.github.io .

翻译：各种适配方法（如LoRA、提示词和适配器）已被提出，用以提升预训练视觉-语言模型在特定领域的性能。然而，这些适配方法在分布偏移下的鲁棒性尚未得到充分研究。本研究评估了11种广泛使用的适配方法在4个视觉-语言数据集上面对多模态干扰时的鲁棒性。具体而言，我们引入了7个基准数据集，涵盖96种视觉干扰和87种文本干扰，旨在探究不同适配方法的鲁棒性、可用适配样本的影响以及适配过程中可训练参数规模的影响。分析揭示：1）适配方法对文本干扰的敏感性高于视觉干扰；2）全参数微调并非始终提供最高鲁棒性，适配器反而能在保持相近干净性能的前提下实现更优鲁棒性；3）与预期相反，增加适配数据量和参数量并不能保证鲁棒性提升，反而导致其下降。我们期望本研究能推动未来鲁棒多模态适配方法的发展。本研究所使用的基准、代码及数据集可通过https://adarobustness.github.io 获取。