Scholar
Pengsheng Guo
Google Scholar ID: O0yBqysAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
532
H-index
7
i10-index
7
Publications
14
Co-authors
3
list available
Publications
Resume (English only)
Academic Achievements
- - Published papers such as 'CAR: Condition-Aware Reparameterization Aligns Source and Target for Better Flow Matching' and multiple other papers in top-tier conferences
- - Presented research findings at NeurIPS, ICCV, ICML, and other leading conferences
- - Involved in patents and technical reports, such as 'Apple's On-Device and Server Foundation Language Models'
Research Experience
- - Tech Lead Manager at Apple, working at the intersection of cutting-edge machine learning research and large-scale product deployment
- - Led the development and launch of the 2-bit on-device Apple Foundation Model, now running on hundreds of millions of iPhones, iPads, and Macs
- - Contributed several core features to Apple Vision Pro, including scene understanding and Persona
Education
- - Worked with Professor David Held on Reinforcement Learning in Robotics during graduate studies at CMU Robotics Institute
- - Worked with Professor I-Ming Chen on Robotic Manipulation during undergrad at NTU, Singapore
Background
- - Tech Lead Manager at Apple
- - Research interests: Multimodal LLM, diffusion model (image, video, 3D), neural rendering, multi-task learning, neural architecture search, and network optimization
