AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Linguistic Binding in Diffusion Models: Enhancing Attribute Correspondence through Attention Map Alignment

Royi Rassin, Eran Hirsch, Daniel Glickman, Shauli Ravfogel, Yoav Goldberg, Gal Chechik

NeurIPS 2023 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 16 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 SynGen syntax-guided attention loss instead of Attend-and-ExciteStable Diffusion 1.4 · DVMPProper Binding: 52.47 -> 74.90Improved
use SynGen instead of Attend-and-ExciteStable Diffusion 1.4 · A&E datasetImproper Binding: 63.81 -> 23.81Improved
use SynGen instead of Attend-and-ExciteStable Diffusion 1.4 · ABC-6Kconcept separation, human majority vote (%): 11.17 -> 28.00Improved
use SynGen instead of Attend-and-ExciteStable Diffusion 1.4 · DVMPvisual appeal, human majority vote (%): 12.17 -> 16.00Improved
use SynGen instead of Attend-and-ExciteStable Diffusion 1.4 · DVMPEntity Neglect: 10.77 -> 16.26Hurt
use SynGen instead of Attend-and-ExciteStable Diffusion 1.4 · ABC-6KEntity Neglect: 33.18 -> 34.41No clear effect
add negative loss term to positive-only loss (ablation)SynGen · DVMP (100 prompts)human preference among 4 variantsImproved
add positive loss term to negative-only loss (ablation)SynGen · DVMP (100 prompts)human preference among 4 variantsImproved
intervene for 25 of 50 steps instead of fewer steps (ablation)SynGen · DVMP (100 prompts)qualitative bindingImproved
intervene for all 50 steps instead of 25 (ablation)SynGen · DVMP (100 prompts)image blur rateHurt
raise scale factor from 1 to 20 (ablation)SynGen · DVMP (100 prompts)qualitative bindingMixed

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