A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training
In this paper, we explore a novel task of Multimodal Unsupervised Continual Post-Training (MU-CPT), enabling deployed MLLMs to continually evolve from streaming unlabeled data. Existing unsupervised post-training methods for MLLMs typically optimize target tokens uniformly, overlooking their heterogeneous visual dependence (VD). However, we reveal that token-level VD is crucial for MU-CPT. Specifically, its structural distortion serves as an indicator of cross-modal catastrophic forgetting, and its inherent heterogeneity acts as a compass to guide new-task learning. Leveraging this property, w
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 59%Exploring continual learning without replay buffers: Our findings using dynamic task-similarity routing [P] →
- FuzzyOverlapping authors or contributors · 62%deepspeedai/DeepSpeed →
“Shared author/contributor keys: lai”
- FuzzyOverlapping authors or contributors · 62%TauricResearch/TradingAgents →
“Shared author/contributor keys: xiao”
- LinkedLinked via arxiv author · 85%Kaichen Li →
“A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training”
- LinkedLinked via arxiv author · 85%Zhilin Zhu →
“A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training”
- LinkedLinked via arxiv author · 85%Jianhao Huang →
“A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training”
- LinkedLinked via arxiv author · 85%Zhengqin Lai →
“A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training”
- LinkedLinked via arxiv author · 85%Baochen Xiong →
“A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training”