On Token's Dilemma: Dynamic MoE with Drift-Aware Token Assignment for Continual Learning of Large Vision Language Models
Paper • 2603.27481 • Published • 29
Mingsong_Li
Mingsong07
AI & ML interests
None yet
Recent Activity
upvoted a paper about 3 hours ago
MHPO: Modulated Hazard-aware Policy Optimization for Stable Reinforcement Learning authored a paper about 18 hours ago
On Token's Dilemma: Dynamic MoE with Drift-Aware Token Assignment for Continual Learning of Large Vision Language Models upvoted a paper about 24 hours ago
HyperAlign: Hypernetwork for Efficient Test-Time Alignment of Diffusion ModelsOrganizations
None yet
authored a
paper about 18 hours ago