Improving Zero-Shot Generalization for CLIP with Synthesized Prompts
This addresses a practical limitation in real-world applications where labeled data is scarce for certain classes, such as emerging concepts, though it is incremental as it builds on existing fine-tuning methods.
The paper tackles the problem of adapting pretrained vision-language models like CLIP to downstream tasks when labeled data is unavailable for some classes, by proposing SHIP, a plug-and-play generative method that synthesizes prompts to generate features for label-only classes, improving zero-shot generalization across various benchmarks.
With the growing interest in pretrained vision-language models like CLIP, recent research has focused on adapting these models to downstream tasks. Despite achieving promising results, most existing methods require labeled data for all classes, which may not hold in real-world applications due to the long tail and Zipf's law. For example, some classes may lack labeled data entirely, such as emerging concepts. To address this problem, we propose a plug-and-play generative approach called \textbf{S}ynt\textbf{H}es\textbf{I}zed \textbf{P}rompts~(\textbf{SHIP}) to improve existing fine-tuning methods. Specifically, we follow variational autoencoders to introduce a generator that reconstructs the visual features by inputting the synthesized prompts and the corresponding class names to the textual encoder of CLIP. In this manner, we easily obtain the synthesized features for the remaining label-only classes. Thereafter, we fine-tune CLIP with off-the-shelf methods by combining labeled and synthesized features. Extensive experiments on base-to-new generalization, cross-dataset transfer learning, and generalized zero-shot learning demonstrate the superiority of our approach. The code is available at \url{https://github.com/mrflogs/SHIP}.