AI Paper Atlas

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

Too Vivid to Be Real? Benchmarking and Calibrating Generative Color Fidelity

Zhengyao Fang, Zexi Jia, Yijia Zhong, Pengcheng Luo, Jinchao Zhang, Guangming Lu, Jun Yu, Wenjie Pei

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 2 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 differentiable softrank loss instead of pairwise loss (ablation)CFM · CFD-Test SynPairsaccuracy (%): 76.2 -> 83.6Improved
use differentiable softrank loss instead of pairwise loss (ablation)CFM · CFD-HumanSpearman: 78.3 -> 84.9Improved
add text conditioning to visual encoder (ablation)CFM · CFD-Test SynPairsaccuracy (%): 77.1 -> 83.6Improved
add text conditioning to visual encoder (ablation)CFM · CFD-Test Real&Synaccuracy (%): 74.3 -> 80.1Improved
add text conditioning to visual encoder (ablation)CFM · CFD-HumanSpearman: 78.9 -> 84.9Improved
add spatial modulation to temporal-only CFR (ablation)CFR on SD3.5 · SD3.5 CFMCFM: -1.3 -> 6.9Improved
add temporal modulation to spatial-only CFR (ablation)CFR on SD3.5 · SD3.5 saturation differenceDelta Sat.: 0.12 -> 0.07Improved
add temporal decay alone to fixed guidance (ablation)CFR on SD3.5 · SD3.5 FIDFID: 13.3 -> 18.0Hurt
add spatial-only modulation to fixed guidance (ablation)CFR on SD3.5 · SD3.5 CFMCFM: 4.9 -> 6.8Improved
use CFM-derived attention instead of HPSv3-derived attention in CFRCFR on SD3.5 · SD3.5 CFMCFM: 5.6 -> 6.9Improved
apply CFR_CFMHunyuan-DiT · Hunyuan FIDFID: 22.1 -> 19.9Improved
apply CFR_CFMPixArt-Sigma · PixArt-Sigma saturation differenceDelta Sat.: 0.09 -> 0.02Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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