DiT-Dec base checkpoint from "CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion", pretrained on LIBERO-90.
CLARE is a general, parameter-efficient framework for exemplar-free continual learning with Vision-Language-Action (VLA) models. It introduces lightweight modular adapters into selected feedforward layers and autonomously expands the model only where necessary when learning a new task, guided by layer-wise feature similarity. During deployment, an autoencoder-based routing mechanism dynamically activates the most relevant adapters without requiring task labels.
1@article{romer2026clare,
2 title={CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion},
3 author={Ralf R{\"o}mer and Yi Zhang and Angela P. Schoellig},
4 journal={arXiv preprint arXiv:2601.09512},
5 year={2026}
6}
