Paper ID: 2407.17491

Robust Adaptation of Foundation Models with Black-Box Visual Prompting

Changdae Oh, Gyeongdeok Seo, Geunyoung Jung, Zhi-Qi Cheng, Hosik Choi, Jiyoung Jung, Kyungwoo Song

With the surge of large-scale pre-trained models (PTMs), adapting these models to numerous downstream tasks becomes a crucial problem. Consequently, parameter-efficient transfer learning (PETL) of large models has grasped huge attention. While PETL methods show impressive performance, they commonly rely on two optimistic assumptions: 1) the entire parameters of a PTM are available, and 2) a sufficiently large memory capacity is equipped for caching all the intermediate activations to compute gradients. However, in most real-world applications, PTMs are served as black-box APIs or proprietary software without explicit parameter accessibility. Besides, it is hard to meet a large memory requirement for modern PTMs. This work proposes black-box visual prompting (BlackVIP), which efficiently adapts the PTMs without knowledge about model architectures and parameters. BlackVIP has two components; 1) Coordinator and 2) simultaneous perturbation stochastic approximation with gradient correction (SPSA-GC). The Coordinator designs input-dependent visual prompts, which allow the target PTM to adapt in the wild. SPSA-GC efficiently estimates the gradient of PTM to update the Coordinator. Besides, we propose a variant, BlackVIP-SE, which significantly reduces the runtime and computational cost of BlackVIP. Extensive experiments on 19 datasets demonstrate that BlackVIPs enable robust adaptation to diverse domains and tasks with minimal memory requirements. We further provide theoretical analysis on the generalization of visual prompting methods by presenting their connection to the certified robustness of randomized smoothing.

Submitted: Jul 4, 2024