Jiwan Hur

SSG: Scaled Spatial Guidance for Multi-Scale Visual Autoregressive Generation

Progressive Latent Calibration for Stable Score Distillation

Efficient Multi-View 3D Representation via Fusion of View-Agnostic Transformations

Inlier-Centric Post-Training Quantization for Object Detection Models

Condensing Videos by Learning Where Motion Matters

Frequency-Aware Token Reduction for Efficient Vision Transformer