AI Paper Atlas

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

GradBias: Unveiling Word Influence on Bias in Text-to-Image Generative Models

Moreno D’Incà, Elia Peruzzo, Massimiliano Mancini, Xingqian Xu, Humphrey Shi, Nicu Sebe

TPAMI 2025 · Image generation › Text-to-image diffusion · approach: Fairness and bias in generative models · 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
use GradBias instead of random word selectionword-influence ranking · Word-influence dataset (SD-1.5)top-1 accuracy: +12.29%Improved
use GradBias instead of random word selectionword-influence ranking · Word-influence dataset (SD-2)top-1 accuracy: +10.70%Improved
use GradBias instead of random word selectionword-influence ranking · Word-influence dataset (SD-XL)top-1 accuracy: +6.71%Improved
use GradBias instead of the second-best baselineword-influence ranking · Word-influence dataset (SD-1.5, SD-2, SD-XL)top-1 accuracy: +9.58%, +3.28%, +2.82%Improved
add image context via Llava-1.5-13B as answer-ranking baselineLLM answer-ranking baseline · Word-influence dataset (SD-2)top-1 accuracy: +7.42%Improved
use Llama3-70B yes/no ranking instead of randomrandom baseline · Word-influence dataset (SD-1.5, SD-XL)top-1 accuracy: +2.71% and +3.64%Improved
use dependency-tree ranking instead of randomrandom baseline · Word-influence dataset (SD-1.5)top-1 accuracy: 36.36 -> 36.86Improved
compute the loss at more than only the final denoising step (every 1, 2, 5 or 10 steps vs only the 50th) (ablation)GradBias · Word-influence dataset (SD-2)top-1 and top-2 accuracy: -4.55% top-1 and -8.84% top-2 when only the 50-th step is usedImproved
swap CLIP-L VQA for Llava-1.5-13B (ablation)GradBias · Word-influence dataset (SD-2)top-1 accuracy: 47.98 -> 48.23No clear effect
swap CLIP-L VQA for Llava-1.5-13B (ablation)GradBias · Word-influence dataset (SD-XL)top-1 accuracy: 44.25 -> 43.22No clear effect
increase number of generated images from N=1 to N=10 (ablation)GradBias · Word-influence dataset (SD-2)top-1 accuracy: 47.47 -> 47.98No clear effect
increase number of generated images from N=1 to N=10 (ablation)GradBias · Word-influence dataset (SD-XL)top-1 accuracy: 42.19 -> 44.25Improved

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