← Search

Chan-Feng Hsu

2 accepted papers

2025

Perspective-Aware Teaching: Adapting Knowledge for Heterogeneous Distillation

ICCV 2025poster

Knowledge distillation (KD) involves transferring knowledge from a pre-trained heavy teacher model to a lighter student model, thereby reducing the inference cost while maintaining comparable effectiveness. Prior KD techniques typically assume homogeneity between the teacher and student models. Howe…

2024

EmoVIT: Revolutionizing Emotion Insights with Visual Instruction Tuning

CVPR 2024poster

Visual Instruction Tuning represents a novel learning paradigm involving the fine-tuning of pre-trained language models using task-specific instructions. This paradigm shows promising zero-shot results in various natural language processing tasks but is still unexplored in vision emotion understandi…