1
Is Our Benchmark Enough? An Analysis of Continual Learning for MLLMs
多模态大模型持续学习基准够用吗？ICML 2026 Workshop论文深度剖析现有评估体系的不足。
arXiv:2606.20961v1 Announce Type: cross Abstract: Continual adaptation is essential for multimodal large language models (MLLMs) deployed across evolv…