Bayesian Principles Improve Prompt Learning In Vision-Language Models
This addresses a key limitation in fine-tuning methods for vision-language models, offering a solution to enhance generalizability, though it is incremental in nature.
The paper tackles overfitting in prompt learning for vision-language models by introducing a Bayesian training objective that balances adaptability and generalizability, resulting in improved performance on target tasks.
Prompt learning is a popular fine-tuning method for vision-language models due to its efficiency. It requires a small number of additional learnable parameters while significantly enhancing performance on target tasks. However, most existing methods suffer from overfitting to fine-tuning data, yielding poor generalizability. To address this, we propose a new training objective function based on a Bayesian learning principle to balance adaptability and generalizability. We derive a prior over the logits, where the mean function is parameterized by the pre-trained model, while the posterior corresponds to the fine-tuned model. This objective establishes a balance by allowing the fine-tuned model to adapt to downstream tasks while remaining close to the pre-trained model.