AI Paper Atlas

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

Diversity-Aware Online Prompt Assignment to Generative Models

Donya Jafari, Farzan Farnia

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Fairness and bias in generative models

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
add diversity term (DAK-UCB) to bandit model selection instead of fidelity-only PAK-UCBonline prompt-to-model assignment · simulated animal image armsselection of the more diverse armImproved
use Mixture-DAK-UCB instead of single-arm DAK-UCBDAK-UCB · MS-COCO prompts (Kandinsky, SDXL, GigaGAN)Joint-RKE and KD scoreImproved
use a mixture of LLMs instead of any single LLMLlama3.2, Qwen2, Gemma3 · city sentence generationCond-Vendi diversityImproved
add diversity term to selection when arms are prompt-specific expertsDAK-UCB · MS-COCO expert armsprompt relevance (CLIP score)No clear effect
use DAK-UCB with I-JRKE for bias detection between unconditioned and capital-conditioned armsDeepSeek · LLM bias detectionselection ratio: 71.5% vs 28.5%Improved

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