Papers › All Roads Lead to Rome? Exploring Representational Similarities Between Latent Spaces...

All Roads Lead to Rome? Exploring Representational Similarities Between Latent Spaces of Generative Image Models

18 Jul 2024arXiv:2407.13449archive 2025-07-28

Charumathi Badrinath, Usha Bhalla, Alex Oesterling, Suraj Srinivas, Himabindu Lakkaraju

Do different generative image models secretly learn similar underlying representations? We investigate this by measuring the latent space similarity of four different models: VAEs, GANs, Normalizing Flows (NFs), and Diffusion Models (DMs). Our methodology involves training linear maps between frozen latent spaces to "stitch" arbitrary pairs of encoders and decoders and measuring output-based and probe-based metrics on the resulting "stitched'' models. Our main findings are that linear maps between latent spaces of performant models preserve most visual information even when latent sizes differ; for CelebA models, gender is the most similarly represented probe-able attribute. Finally we show on an NF that latent space representations converge early in training.

PaperPDFCode

Code

charumathib/thesis-latent-spaces officialmentioned in paperjax report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

AllAttribute

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

DiffusionNormalizing Flows

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections