AI Paper Atlas

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

FairSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation

Tatiana Gaintseva, Akshit Achara, Greg Slabaugh, Jiankang Deng, Ismail Elezi

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance

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
apply EquiSteer cross-attention steering for gender debiasingSD-1.5 · Gender, neutral prompts (8 professions)avg parity gap Delta: 0.403 -> 0.051Improved
EquiSteer vs training-free TEI baselineSD-1.5 · Gender, neutral promptsavg parity gap Delta: 0.167 -> 0.051Improved
add orthogonalisation against attribute directions to adaptive steering (ablation)EquiSteer · SD-1.5 gender, neutral promptsneutral Delta: 0.165 -> 0.095Improved
add orthogonalisation without gate (ablation)EquiSteer · SD-1.5 gender, male-specific promptsmale-specific ratio: 0.379 -> 0.230Hurt
add prompt-aware gate (ablation)EquiSteer (add & erase) · SD-1.5 genderneutral Delta / male-specific ratio: neutral 0.095 -> 0.051; male-specific 0.230 -> 0.495Improved
apply EquiSteer to race debiasingSDXL and SANA · Race, neutral promptsparity gap Delta: 0.172 -> 0.042 (SDXL), 0.178 -> 0.039 (SANA)Improved
apply EquiSteer to eyeglassesSANA · Eyeglasses, neutral prompts (BLIP-VQA)parity gap Delta: 0.234 -> 0.175Improved
debias four concepts jointly in one passEquiSteer · Joint-4 gender+race+age+body typeparity gap Delta: gender SDXL 0.381 -> 0.012; body type SANA 0.391 -> 0.178Improved
apply EquiSteer to compositional/multi-subject promptsSANA · Gender, compositional promptsparity gap Delta: 0.295 -> 0.155Improved
EquiSteer vs FairImagen on FairImagen setupSDXL · Gender, pooled Delta (10 steps)pooled Delta: 0.098 -> 0.008Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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