Mattia Soldan

long-form video (2) contrastive learning (1) vision transformer (1) action recognition (1) self-attention mechanism (1) transfer learning (1) video generation (1) knowledge distillation (1) multimodal learning (1) egocentric vision (1) multi-modal learning (1) temporal grounding (1) autoregressive model (1) benchmark dataset (1) video-text retrieval (1) encoder-decoder architecture (1) encoder-decoder model (1) natural language grounding (1) video grounding (1) token reduction (1)