AI Paper Atlas

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

GenExam: A Multidisciplinary Text-to-Image Exam

Zhaokai Wang, Penghao Yin, Xiangyu Zhao, Changyao Tian, Yu Qiao, Wenhai Wang, Jifeng Dai, Gen Luo

ICML 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters

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 ground-truth image as evaluator input (ablation)GPT-5 low evaluator · GenExam evaluator ablationscoring point accuracy (%): 86.2 -> 88.5Improved
give ground-truth text caption instead of image (ablation)GPT-5 low evaluator · GenExam evaluator ablationscoring point accuracy (%): 88.5 -> 87.8Hurt
add text caption on top of ground-truth image (ablation)GPT-5 low evaluator · GenExam evaluator ablationscoring point accuracy (%): 88.5 -> 88.5No clear effect
raise reasoning effort from minimal to low (ablation)GPT-5 evaluator · GenExam evaluator ablationscoring point accuracy (%): 84.2 -> 88.5Improved
raise reasoning effort from low to medium (ablation)GPT-5 evaluator · GenExam evaluator ablationscoring point accuracy (%): 88.5 -> 86.7Hurt
raise reasoning effort from low to high (ablation)GPT-5 evaluator · GenExam evaluator ablationscoring point accuracy (%): 88.5 -> 87.6Hurt
raise reasoning effort from minimal to low (ablation)GPT-5-mini evaluator · GenExam evaluator ablationscoring point accuracy (%): 79.0 -> 84.3Improved
use GPT-5 low instead of GPT-4o as evaluator (ablation)evaluator choice · GenExam evaluator ablationscoring point accuracy (%): 68.5 -> 88.5Improved
enable thinking modeBAGEL · GenExam overallrelaxed score (%): 11.4 -> 12.9Improved
use dedicated T2I model instead of unified MLLMFLUX.2 dev vs BAGEL · GenExam overallrelaxed score (%): 11.4 -> 42.3Improved
closed-source instead of open-source modelNano Banana Pro vs FLUX.2 dev · GenExam overallstrict score (%): 2.4 -> 70.2Improved

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