Research breakthrough · curated entry
DDPM makes diffusion the main generative line
Denoising Diffusion Probabilistic Models
- Date
- June 19, 2020
- Category
- Research breakthrough
- Organization / Model
- University of California, Berkeley
- Evidence
- 1 source · 1 evidence family
Ho, Jain and Abbeel proposed denoising diffusion probabilistic models: gradually noise an image, then train a network to reverse the process and sample from pure noise. It matched the best GANs of the day on image quality and later became shared infrastructure for Stable Diffusion, DALL-E 2 and video generation.
Curator's note: Cross-checked against the arXiv preprint; the paper was published at NeurIPS in 2020.
Curated sources · 1
Cite this page
This page is a curated static snapshot and can be cited directly. Preferred format GB/T 7714—2025 (electronic resource [EB/OL]); the other 5 styles open below, together with the machine-readable files.
GB/T 7714—2025 (electronic resource [EB/OL])
AI 纪年. DDPM 提出，扩散模型成为生成模型主线[EB/OL]. (2020-06-19)[2026-11-03]. https://aihist.net/e/ddpm-2020.html.
APA 7 · American Psychological Association
AI 纪年. (2020, June 19). DDPM 提出，扩散模型成为生成模型主线 [Web page]. https://aihist.net/e/ddpm-2020.html
MLA 9 · Modern Language Association
"DDPM 提出，扩散模型成为生成模型主线." AI 纪年, 19 June 2020, aihist.net/e/ddpm-2020.html. Accessed 3 Nov. 2026.
Chicago 17 · notes-bibliography
AI 纪年. "DDPM 提出，扩散模型成为生成模型主线." Published June 19, 2020. Accessed Nov. 3, 2026. https://aihist.net/e/ddpm-2020.html.
IEEE · citation-reference list
"DDPM 提出，扩散模型成为生成模型主线," AI 纪年, 19 June 2020. [Online]. Available: https://aihist.net/e/ddpm-2020.html. [Accessed: 3 Nov. 2026].
ACM · author-date variant
AI 纪年. June 19, 2020. DDPM 提出，扩散模型成为生成模型主线. Retrieved November 3, 2026 from https://aihist.net/e/ddpm-2020.html.
All six styles come from one template and can be compared word for word. [引用日期], Accessed, urldate and Y2 all carry this page's build date, 2026-11-03 — a static page is an immutable snapshot, so substitute the date you actually consulted it. The address in every style is the archive's canonical page: this English page translates the same snapshot, and the work keeps its name, so one record cites both. For one-click copying use the same entry in the timeline, which takes the day you opened it. Expand the other styles before printing.