@Y0Yunpeng’s Homepage
🧐 About Me
Hi, I'm Yunpeng, an AI practitioner passionate about pushing the boundaries of large model capabilities. My research focuses on vision-language models, spatial reasoning, and embodied AI. Beyond that, my industry experience centers on Application & Post-training of LLMs, VLMs, especially in RL/RLVR
I'm committed to turning research ideas into practical, reproducible engineering work. I believe AI is not a weapon to replace humans, but to reduce repetitive work and amplify productivity. People can spend more time on creativity, exploration, judgment, or pursue some values uniquely human.
Open to discussion and collaboration on AI-related projects — feel free to reach out!
📄 CV
📰 Blog Posts
📝 Publications
📖 Education
The Grainger College of Engineering | M.Eng. | Major in Electrical and Computer Engineering
Liangxin College | B.Eng. (Honors) | Major in Optoelectronic Information Science and Engineering
🏆 Honors and Awards
- 2025 | Provincial Government Scholarship
💻 Internships
2026.02 - 2026.07 | ByteDance – Data-AML
VLM/LLM Application Algorithm Intern • Shenzhen, Guangdong, China
2025.10 - 2026.01 | XPeng Motors – Intelligent Cockpit Center
VLM Foundation Model Algorithm Intern • Beijing, China
2025.07 - 2025.10 | Truesight – AI Research
Computer Vision R&D Intern • Xiamen, Fujian, China
🎨 Portfolio
3DMM-FLUX: Instruction-Guided Facial Expression Editing
An instruction-driven face editing framework that combines interpretable 3DMM expression control with FLUX diffusion generation for identity-preserving facial expression editing.
FacePairEmoji
A multi-resolution paired facial expression dataset for FLUX fine-tuning, built around same-person expression pairs and resolution buckets from 256px to 2048px.
Hobbies
🏋️♂️ Fitness • 🏀 Basketball • 🏊♂️ Swimming • 🏄♂️ Surfing • ⛳ Golf • 🎾 Tennis