Scholar
Yushi Huang
Google Scholar ID: 9-FYp3EAAAAJ
Hong Kong University of Science and Technology
Citations & Impact
Citations
107
H-index
5
i10-index
2
Publications
9
Co-authors
3
list available
Publications
Resume (English only)
Academic Achievements
- June 2025: Paper 'Temporal Feature Matters' accepted to IEEE TPAMI
- May 2025: Paper 'HarmoniCa' accepted to ICML
- October 2024: Paper 'LLMC' accepted to EMNLP Industry Track
- July 2024: Paper 'PTSBench' accepted to ACM MM
- February 2024: Paper 'TFMQ-DM' accepted to CVPR as a Highlight Poster (Top 2.8%)
- Co-authored LinVideo: a data-free post-training framework for efficient video generation, achieving 1.25–2.00× speedup and 15.92× latency reduction with minimal quality drop in distilled 4-step model
- Co-developed LLMC+: a plug-and-play benchmark toolkit for vision-language model compression, supporting over 20 algorithms across five VLM families
