Exploring Visual Prompts for Adapting Large-Scale Models
This provides a new perspective on adapting pre-trained models in vision, though it is incremental as it builds on existing prompt tuning and adversarial reprogramming methods.
The paper tackles adapting large-scale vision models to new tasks using visual prompting, learning a single image perturbation to prompt frozen models, and shows it is effective for CLIP and robust to distribution shift, achieving performance competitive with linear probes.
We investigate the efficacy of visual prompting to adapt large-scale models in vision. Following the recent approach from prompt tuning and adversarial reprogramming, we learn a single image perturbation such that a frozen model prompted with this perturbation performs a new task. Through comprehensive experiments, we demonstrate that visual prompting is particularly effective for CLIP and robust to distribution shift, achieving performance competitive with standard linear probes. We further analyze properties of the downstream dataset, prompt design, and output transformation in regard to adaptation performance. The surprising effectiveness of visual prompting provides a new perspective on adapting pre-trained models in vision. Code is available at http://hjbahng.github.io/visual_prompting .