Convolutional Prompting meets Language Models for Continual Learning


Continual Learning (CL) enables machine learning models to learn from continuously shifting new training data in absence of data from old tasks. Recently, pre-trained vision transformers combined with prompt tuning have shown promise for overcoming catastrophic forgetting in CL. These approaches rely on a pool of learnable prompts which can be inefficient in sharing knowledge across tasks leading to inferior performance. In addition, the lack of fine-grained layer specific prompts does not allow these to fully express the strength of the prompts for CL. We address these limitations by proposing ConvPrompt, a novel convolutional prompt creation mechanism that maintains layer-wise shared embeddings, enabling both layer-specific learning and better concept transfer across tasks. The intelligent use of convolution enables us to maintain a low parameter overhead without compromising performance. We further leverage Large Language Models to generate fine-grained text descriptions of each category which are used to get task similarity and dynamically decide the number of prompts to be learned. Extensive experiments demonstrate the superiority of ConvPrompt which improves the state-of-the-art with significantly less parameter overhead.

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2024
comments powered by Disqus