cs.CVNov 16, 2026

LEDGE: Separating Critic and Editor Failures in Text-to-Image Self-Correction

Authors: Vyom Agarwal, Mokshda Gangrade, Samet Ayhan, Giacinto Paolo Saggese

Organizations: University of Maryland, College Park

Abstract

Text-to-image (T2I) self-correction combines a vision-language model (VLM) critic that diagnoses errors and proposes instructions with an image editor that executes them. End-to-end evaluation obscures whether failures arise from incorrect diagnoses or an editor unable to execute valid instructions. We introduce LEDGE (Learning Error Diagnosis in Generation and Editing), a framework that separates these bottlenecks. LEDGE compiles oracle edit instructions from GenEval while holding the generated images and the editor fixed, isolating the editor's correction capability from the critic's diagnostic accuracy. On Stable Diffusion 3.5 Large, oracle instructions raise GenEval from 72.0% to 81.2%, versus 74.1% for the baseline VLM critic. Yet 435 of 644 failed images remain unrepaired, exposing a substantial editor gap. We also use oracle labels from disjoint training prompts to supervise the critic. Oracle supervision raises exact diagnosis from 33.3% (the same VLM with our structured output, untrained) to 70.8%. With the same editor, the resulting critic improves GenEval from 42.2% (uncorrected) to 59.4% on Stable Diffusion 1.5, against 50.9% for the baseline critic, and from 72.0% to 76.0% on Stable Diffusion 3.5 Large, despite training only on Stable Diffusion 1.5. Improvements transfer to unseen object classes, a second editor, and independent T2I-CompBench judges and prompts. A small human evaluation confirms the ranking in estimated net corrections but suggests a smaller gain than detector-based evaluation. LEDGE thus distinguishes two limits of self-correction: knowing what to fix and being able to fix it.

Figures & tables

Appendix figures & tables11 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Self-Generated Error Training for Token Editing in Diffusion Language Models

    Jun 15, 2026Lin YaoLLM Self-CorrectionDiffusion Model Fine-Tuning

  2. Self-Corrected Image Generation with Explainable Latent Rewards

    Mar 26, 2026Yinyi Luo, Hrishikesh Gokhale, Marios Savvides +2Image GenerationControllable Image Generation

  3. Where, What, Why, and Importance: Structured Defect Grounding for Text-to-Image Feedback

    Jun 4, 2026Huaisong Zhang, Hao Yu, Yuxuan Zhang +7Diffusion Model AlignmentT2I Generation Evaluation