AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Logical Guidance for the Exact Composition of Diffusion Models

Francesco Alesiani, Jonathan H Warrell, Tanja Bien, Henrik Christiansen, Matheus Ferraz, Mathias Niepert

ICML 2026 · Image generation › Fast diffusion sampling · approach: Guidance for conditional diffusion · 0 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
use probability-weighted negation instead of constant-weight negationLogDiff · CelebA (NOT)FID: 19.7 -> 12.1Improved
use probability-dependent weights instead of constant weights for negationLogDiff · CMNIST (NOT, higher guidance scale)entropy / diversityImproved
add adaptive repulsive guidance (A and not B) (ablation)LogDiff · CMNIST (AND)conformity score (%): 80.4 -> 83.6Improved
add adaptive repulsive guidance (A and not B) (ablation)LogDiff · Shapes3D (AND)conformity score (%): 82.4 -> 86.4Improved
add adaptive repulsive guidance (A and not B) (ablation)LogDiff · CMNIST (OR-ME)conformity score (%): 98.0 -> 97.9No clear effect
add adaptive repulsive guidance (A and not B) (ablation)LogDiff · Shapes3D (N=2..5 operators)conformity score (%): 88.8 -> 91.8 (N=2); 87.6 -> 89.9 (N=5)Improved
add adaptive repulsive guidance with moderate weight (ablation)EDM2-XXL on ImageNet-512 · ImageNet-512FIDImproved
use LogDiff composition instead of Skreta et al. (2025b)diffusion composition · CMNIST and Shapes3D ANDconformity score (%): 26.2 -> 80.4 (CMNIST); 30.1 -> 82.4 (Shapes3D)Improved
use LogDiff negation instead of Liu et al. (2022) constant-weight negationLogDiff · CMNIST and Shapes3D (NOT)conformity score (%): 99.5 -> 96.5 (CMNIST); 98.7 -> 96.4 (Shapes3D)Hurt
add Feynman-Kac Correction to LogDiff (ablation)LogDiff · GRM5-RRM1 A AND Bproduct docking score (A*B): 73.20 -> 80.89Improved
add Feynman-Kac Correction to LogDiff (ablation)LogDiff · GRM5-RRM1 A OR Bproduct docking score (A*B): 73.91 -> 73.51Hurt
add Feynman-Kac Correction to LogDiff (ablation)LogDiff · GRM5-RRM1 molecule diversity and validitydiversity: 0.89 -> 0.73 (AND)Hurt

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