SourceThe cases citing it, and what each argues, are listed below.
A Reproducible Extraction of Training Images from Diffusion Models
arXiv.orgNot provided
- Source type
- Academic Paper
- Published
- Not provided
- Added by
- PsamiKoinovera team
- Added
- 6 Oct 2026
Contributor evidence
What this source establishes
Reports reproducible extraction across image-generation systems. Provides additional examples of memorisation beyond a single model study.
Limitations
Targeted extraction and older models; reproduction frequency in everyday use remains separate.
Supporting excerpt
Carlini et al. demonstrated the widely used model Stable Diffusion can regurgitate real training samples
AI evidence analysis
Koinovera can compare extracted source evidence against this claim. The result is advisory and may be wrong.
Sign in to analyse evidenceHow readers rated this source
These rate the source itself, wherever it is cited. The reasoning built on it is rated on each case.
No one has rated this source yet.
Sign in to rate this source