Methodologies for diffusion model interpretability: A systematic review

Lakhani, Tina, Wu, Jing ORCID: https://orcid.org/0000-0001-5123-9861, Lai, Yukun ORCID: https://orcid.org/0000-0002-2094-5680 and Ji, Ze ORCID: https://orcid.org/0000-0002-8968-9902 2025. Methodologies for diffusion model interpretability: A systematic review. IEEE Transactions on Artificial Intelligence 10.1109/TAI.2025.3648376

PDF - Accepted Post-Print Version Available under License Creative Commons Attribution. Download (526kB)

Abstract

Diffusion generative models have gained rapid traction since 2020 due to their expressiveness and highquality outputs. Explaining and interpreting these models is essential for enabling further improvements and fostering trustworthiness. This systematic review identifies and analyzes interpretability methods applied to diffusion models across domains, highlighting key trends, outlining strategies, and identifying emerging research directions. We screened 1,489 papers published between 2020–2025 across IEEE, Scopus, DBLP, arXiv, and Elicit, and included 81 studies that met predefined criteria. Most methods target latent space analysis (n = 35), followed by data attribution (n = 16) and denoising dynamics (n = 14). Image generation and text-to-image synthesis dominate application areas (n = 73), with limited coverage in robotics, audio, and neuroscience (n = 8). This review offers a structured taxonomy, quantifies interpretability research trends, and identifies domain–specific and architectural gaps.

Item Type:	Article
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Computer Science & Informatics Schools > Engineering
Additional Information:	RRS policy applied
Date of First Compliant Deposit:	18 March 2026
Date of Acceptance:	13 December 2025
Last Modified:	18 Mar 2026 11:45
URI:	https://orca.cardiff.ac.uk/id/eprint/185836

Actions (repository staff only)

Edit Item