Shanshan Zhong

large language model (4) benchmark evaluation (3) model selection (1) self-attention mechanism (1) data compression (1) multimodal learning (1) neural network optimization (1) mixture of expert (1) hierarchical alignment (1) instruction tuning (1) video understanding (1) parametric model (1) dynamical system (1) cost-aware learning (1) multi-task learning (1) multimodal large language model (1) model routing (1) ordinary differential equation (1) temporal coherence (1) 3d reconstruction (1)