Theoretical analysis shows DDIM reverse dynamics get stuck on mode-connecting segments in Gaussian mixtures after time τ, while DDPM stochasticity prevents this and lowers hallucination rates.
Laplacian score sharpening for mitigating hallucination in diffusion models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse Dynamics
Theoretical analysis shows DDIM reverse dynamics get stuck on mode-connecting segments in Gaussian mixtures after time τ, while DDPM stochasticity prevents this and lowers hallucination rates.