Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse Dynamics
On the map: Diffusion models → Medical imaging and segmentation
Abstract (arXiv)
We theoretically study the hallucination phenomena in two canonical diffusion samplers: the stochastic Denoising Diffusion Probabilistic Model (DDPM) and the deterministic Denoising Diffusion Implicit Model (DDIM). We analyze the reverse ODE (DDIM) and SDE (DDPM) for a Gaussian mixture target, proving that after a critical time τ, (a) DDIM can become stuck on the segment connecting the two nearest modes and (b) DDPM *stochasticity* helps it become unstuck from this region, thus avoiding hallucination. Our empirical validation verifies that DDPM has a significantly lower hallucination rate than DDIM when this region is entered. Building on our observations, we exhibit how using additional stochastic steps can help DDIM avoid hallucinations and offer new insights on how to design improved samplers.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 6 of its 41 references that are in the map. First posted on arXiv 7 May 2026.
Builds on
6 of 41 references.
Closest ancestors · 3
Same direction, newest first
- 2025Laplacian Score Sharpening for Mitigating Hallucination in Diffusion Models10 Nov
- 2025Counting Hallucinations in Diffusion Models15 Oct
- 2025Mitigating Diffusion Model Hallucinations with Dynamic Guidance6 Oct
Other references · 3
- 2025Minimax Optimality of the Probability Flow ODE for Diffusion Modelscited by 23
- 2025Dimension-Free Convergence of Diffusion Models for Approximate Gaussian Mixturescited by 23
- 2025Convergence of Deterministic and Stochastic Diffusion-Model Samplers: A Simple Analysis in Wasserstein Distancecited by 21
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2024Understanding Hallucinations in Diffusion Models through Mode Interpolation
- 2024Denoising Diffusion Probabilistic Models in Six Simple Steps
- 2026Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
- 2023Fast Diffusion Probabilistic Model Sampling through the lens of Backward Error Analysis
- 2026A Tutorial on Diffusion Theory: From Differential Equations to Diffusion Models
- 2026Score-Control for Hallucination Reduction in Diffusion Models
- 2026Lost in Diffusion: Uncovering Hallucination Patterns and Failure Modes in Diffusion Large Language Models
- 2026Connections between the F"ollmer process and the denoising diffusion probabilistic model