AI Paper Atlas

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

FailureAtlas: Mapping the Failure Landscape of T2I Models via Active Exploration

Muxi Chen, Zhaohua Zhang, Chenchen Zhao, Mingyang Chen, Wenyu Jiang, Tianwen Jiang, Jianhuan Zhuo, Yu Tang, Qiuyong Xiao, Jihong Zhang, Qiang Xu

CVPR 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 rule-based pruning that skips extensions of failed nodes (ablation)FailureAtlas tree search · SD1.5, layer 3search space size: reduced to 4.2% at layer 3, 0.4% at extrapolated layer 4Improved
add prediction-based prioritization of likely-failing nodes (ablation)FailureAtlas tree search · SD1.5, layer 3failure discovery speed: roughly 2x speed-upImproved
pruning applied to a weaker model (SD1.5 vs SDXL Turbo)rule-based pruning · SD1.5 vs SDXL Turbosearch space reduction: 439,467 vs 1,040,415 explored nodesImproved
use online-trained predictor of node success rateprioritizer · SD1.5 node success predictionL1 loss on unseen nodesImproved
increase repeated generations per node beyond 25 (ablation)node success rate estimate · SD1.5 and SDXL Turboaverage change in success rate: less than 0.6% average changeNo clear effect
low training-data frequency of an entity or attributeSD1.5 error slices · LAION-2B-en, layer 1error attribution to data scarcityMixed
add attributes to an entity prompt (layer 1 to layer 2)SD1.5 and SDXL Turbo · error density by layererror density: SD1.5 21.3% -> 84.1%; SDXL Turbo 12.0% -> 72.0%Hurt

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