Alejandro Pardo

video generation (2) contrastive learning (1) self-attention mechanism (1) weakly supervised learning (1) multimodal learning (1) multi-modal learning (1) temporal grounding (1) iterative refinement (1) autoregressive model (1) benchmark dataset (1) video editing (1) long-form video (1) training-free method (1) encoder-decoder architecture (1) temporal action localization (1) encoder-decoder model (1) shot selection (1) text-to-video diffusion (1) pseudo ground truth (1) joint diffusion (1)