Scholar
Po-Yao (Bernie) Huang
Google Scholar ID: E8K25LIAAAAJ
FAIR, Meta
Citations & Impact
Citations
24,482
H-index
30
i10-index
54
Publications
20
Co-authors
18
list available
Publications
Cited
77
Resume (English only)
Academic Achievements
- Contributor to the Llama 3.2 multimodal component as part of the Llama 3 Team
- Chameleon: Mixed-modal early-fusion foundation models (Joint first author; contributed to pre-training)
- Cm3: A causal masked multimodal model of the internet
- DINOv2: Learning robust visual features without supervision (JMLR, 2024)
- VoiceCraft: Zero-Shot Speech Editing and Text-to-Speech in the Wild (ACL, 2024)
- SeamlessM4T—Massively Multilingual & Multimodal Machine Translation (Nature, 2024)
- Altogether: Image Captioning via Re-aligning Alt-text (EMNLP, 2024)
- Demystifying CLIP Data (MetaCLIP) (ICLR, 2024)
- MoDE: CLIP Data Experts via Clustering (CVPR, 2024)
- MAViL: Masked Audio-Video Learners (NeurIPS, 2023)
- Diffusion Models as Masked Autoencoders (ICCV, 2023)
- CiT: Curation in Training for Effective Vision-Language Data (ICCV, 2023)
- Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles
Research Experience
- Facebook (Meta): Senior Research Scientist, FAIR Labs (Aug 2022 – present)
- Facebook (Meta): Research Scientist, FAIR Labs (Aug 2021 – Aug 2022)
- Facebook (Meta): Research Intern (May 2020 – May 2021)
- Microsoft: Research Intern, Microsoft Research (Jun 2017 – Aug 2017)
- MediaTek: Senior Software Engineer (Jun 2012 – Jun 2014)
- MediaTek: Software Engineer (Sep 2010 – May 2012)
