Scholar
Jiaming Song
Google Scholar ID: 6dP660cAAAAJ
Stanford University
Machine Learning
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
26,407
H-index
48
i10-index
70
Publications
20
Co-authors
18
list available
Contact
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
Publications
8 items
Terminal Velocity Matching
2025
Cited
0
SABER: Switchable and Balanced Training for Efficient LLM Reasoning
2025
Cited
0
Self-NPO: Negative Preference Optimization of Diffusion Models by Simply Learning from Itself without Explicit Preference Annotations
2025
Cited
0
Inductive Moment Matching
2025
Cited
1
Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms
2025
Cited
0
Personalized Preference Fine-tuning of Diffusion Models
2025
Cited
0
Decentralized Diffusion Models
2025
Cited
0
Score-based Diffusion Models in Function Space
arXiv.org · 2023
Cited
35
Resume (English only)
Academic Achievements
Created the earliest accelerated algorithm for diffusion models, widely used in DALL-E 2, Imagen, Stable Diffusion, and ERNIE-ViLG 2.0.
Co-authored the foundational paper for Stable Diffusion’s img2img method.
Democratized diffusion models for general applications such as super-resolution, inpainting, and JPEG restoration.
Organized the CVPR 2023 diffusion model tutorial and SIGGRAPH 2023 Course on Diffusion Models.
Gave talks at University of Edinburgh, DeepMind, Imperial College London, and Oxford.
First-authored papers at ICLR 2023 and ICML 2023; contributed to multiple CVPR and other top-tier conference publications.
Co-authors
18 total
Stefano Ermon
Stanford University
Chenlin Meng
Stanford University
Shengjia Zhao
OpenAI
Aditya Grover
Co-founder, Inception | Prof, UCLA | PhD, Stanford
Yunzhu Li
Columbia University
Lantao Yu
Ph.D., Computer Science, Stanford University
Co-author 7
Arianna Yuan
Stanford University
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up