Semantic Prompting with Image-Token for Continual Learning

arXiv:2403.11537v1 Announce Type: cross
Abstract: Continual learning aims to refine model parameters for new tasks while retaining knowledge from previous tasks. Recently, prompt-based learning has emerged to leverage pre-trained models to be prompted to learn subsequent tasks without the reliance on the rehearsal buffer. Although this approach has demonstrated outstanding results, existing methods depend on preceding task-selection process to choose appropriate prompts. However, imperfectness in task-selection may lead to negative impacts on the performance particularly in the scenarios where the …

abstract arxiv continual cs.cv cs.lg image knowledge learn parameters pre-trained models prompt prompt-based learning prompting refine reliance results semantic tasks token type