AI 中文总结
综述视觉领域持续自监督学习（CSSL），分析现有评估协议问题，探讨自监督目标抗灾难性遗忘原因，基于遗忘缓解策略对方法分类，识别如可扩展性等挑战，提出推进CSSL需转向大规模持续预训练范式。
AI 中文摘要
传统上，持续学习假设可获取有标签数据，但许多现实世界应用，如终身机器人技术，要求模型从无标签流中持续适应。这促使了持续自监督学习（CSSL）的发展，该领域发展迅速但缺乏专门系统综述。本文对视觉领域的CSSL进行全面综述，并关联新兴视觉语言设置。首先分析现有评估协议，强调阻碍公平比较的不一致性；接着探讨自监督目标对灾难性遗忘有更强鲁棒性的原因；然后基于遗忘缓解策略将现有方法统一分类；最后识别开放性挑战。我们认为推进CSSL需从小规模基准转向大规模系统的持续预训练范式。
英文摘要
Traditionally, continual learning has assumed access to labeled data, yet many real-world applications -- such as lifelong robotics -- require models to adapt continuously from unlabeled streams. This has led to the development of continual self-supervised learning (CSSL), a rapidly growing area that lacks a dedicated, systematic review. In this work, we present a comprehensive survey of CSSL for vision, with connections to emerging vision-language settings. First, we analyze existing evaluation protocols and highlight inconsistencies that hinder fair comparison. We then examine why self-supervised objectives exhibit improved robustness to catastrophic forgetting, relating this to task-agnostic representations and smoother loss landscapes. Next, we organize existing methods into a unified taxonomy based on their forgetting-mitigation strategies, including distillation, replay, regularization, architectural approaches, model merging, and objective-level adaptation. Finally, we identify open challenges such as scalability and the need for fast adaptability. We argue that advancing CSSL requires moving beyond small-scale benchmarks towards continual pre-training paradigms for large-scale systems.
CommentsThis work has been accepted for publication in IEEE EAIS 2026
