Towards Difficulty-Agnostic Efficient Transfer Learning for Vision-Language Models
This work addresses the problem of adapting vision-language models to diverse downstream tasks with varying difficulty levels, offering an incremental improvement over existing efficient transfer learning methods.
The paper tackles the challenge of varying transfer difficulty in efficient transfer learning for vision-language models, proposing an adaptive ensemble method that combines visual prompts and text adapters with pre-trained models, which consistently outperforms baselines across benchmarks, especially on unseen tasks.
Vision-language models (VLMs) like CLIP have demonstrated remarkable applicability across a variety of downstream tasks, including zero-shot image classification. Recently, the use of prompts or adapters for efficient transfer learning (ETL) has gained significant attention for effectively adapting to downstream tasks. However, previous studies have overlooked the challenge of varying transfer difficulty of downstream tasks. In this paper, we empirically analyze how each ETL method behaves with respect to transfer difficulty. Our observations indicate that utilizing vision prompts and text adapters is crucial for adaptability and generalizability in domains with high difficulty. Also, by applying an adaptive ensemble approach that integrates task-adapted VLMs with pre-trained VLMs and strategically leverages more general knowledge in low-difficulty and less in high-difficulty domains, we consistently enhance performance across both types of domains. Based on these observations, we propose an adaptive ensemble method that combines visual prompts and text adapters with pre-trained VLMs, tailored by transfer difficulty, to achieve optimal performance for any target domain. Upon experimenting with extensive benchmarks, our method consistently outperforms all baselines, particularly on unseen tasks, demonstrating its effectiveness.