AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Uncovering Conceptual Blindspots in Generative Image Models Using Sparse Autoencoders

Maty Bohacek, Thomas Fel, Maneesh Agrawala, Ekdeep Singh Lubana

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Classical generative image synthesis · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
compare generated vs real LAION images in SAE concept space across four modelsSD 1.5, SD 2.1, PixArt, Kandinsky · Energy difference distribution over 32,000 RA-SAE concepts (10,000 LAION pairs)energy-difference skewness: skewness -0.54 (SD 2.1), -0.40 (SD 1.5, PixArt), -0.23 (Kandinsky)Mixed
compare SD 2.1 with SD 1.5 on concept blindspot profileSD 2.1 vs SD 1.5 · Cross-model concept blindspot correlationPearson r: r = 0.82No clear effect
compare PixArt and Kandinsky with SD 1.5 on concept blindspot profilePixArt, Kandinsky · Cross-model concept blindspot correlationPearson r: r = 0.41 (PixArt), r = 0.46 (Kandinsky)Mixed
apply DPO fine-tuning to SD 1.5SD 1.5 · Datapoint-wise concept energy differenceL2 norm of concept energy difference: lower median and narrower spread (no numbers given)Improved
higher natural frequency of a conceptSD 1.5/2.1, PixArt, Kandinsky · Concept frequency vs absolute energy differenceenergy discrepancy: higher-frequency concepts show lower energy discrepancies (no numbers given)Improved
use alpha-masked image instead of heatmap overlay for VLM concept description (ablation)GPT-4o autointerpretation · Qualitative concept descriptiondescription correctnessImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy