20 papers
SAFE-Merge: Data-Free Continual Model Merging with General Knowledge Preservation
Zihuan Qiu, Zhiyang Liao, Chiyuan He +5
Data-free continual model merging must incorporate a stream of specialized models while retaining both pretrained general knowledge and previously acquired tasks, without access to…
FutureRTC: Real-Time Robot Execution with Anticipatory-Conditioned Action Chunking
Hai Jiang, Yixian Zou, Binbin Liang +3
Real-time deployment of Vision-Language-Action (VLA) policies necessitates asynchronous execution, wherein subsequent action chunks are computed concurrently with the execution of…
Test-time Ego-Exo-centric Adaptation for Action Anticipation via Multi-Label Prototype Growing and Dual-Clue Consistency
Zhaofeng Shi, Heqian Qiu, Lanxiao Wang +4
Efficient adaptation between Egocentric (Ego) and Exocentric (Exo) views is crucial for applications such as human-robot cooperation. However, the success of most existing Ego-Exo…
Continual Learning with Vision-Language Models via Semantic-Geometry Preservation
Chiyuan He, Zihuan Qiu, Fanman Meng +4
Continual learning of pretrained vision-language models (VLMs) is prone to catastrophic forgetting, yet current approaches adapt to new tasks without explicitly preserving the cros…
DesCLIP: Robust Continual Learning via General Attribute Descriptions for VLM-Based Visual Recognition
Chiyuan He, Zihuan Qiu, Fanman Meng +3
Continual learning of vision-language models (VLMs) focuses on leveraging cross-modal pretrained knowledge to incrementally adapt to expanding downstream tasks and datasets, while…
Null-Space Filtering for Data-Free Continual Model Merging: Preserving Stability, Promoting Plasticity
Zihuan Qiu, Lei Wang, Yang Cao +7
Data-free continual model merging (DFCMM) aims to fuse independently fine-tuned models into a single backbone that evolves with incoming tasks without accessing task data. This pap…