AI Paper Atlas

Research map / Image generation / Image restoration

See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data Synthesis

Jaehyun Park, Minyoung Ahn, Minkyu Kim, Jonghyun Lee, Jae-Gil Lee, Dongmin Park

CVPR 2026 · Image generation › Image restoration · approach: Diffusion-based universal restoration · 0 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
fine-tune on ArtiAgent dataInternVL3.5-8B · ArtiBench detectionaccuracy: 0.498 -> 0.630Improved
fine-tune on ArtiAgent dataQwen2.5-VL-7B · ArtiBench detectionaccuracy: 0.501 -> 0.627Improved
fine-tune on ArtiAgent dataQwen2.5-VL-7B · ArtiBench localizationmIoU: 0.010 -> 0.111Improved
fine-tune on ArtiAgent dataQwen2.5-VL-7B · ArtiBench explanationROUGE: 0.117 -> 0.233Improved
ArtiAgent-tuned open VLM vs GPT-5 on RichHF localizationQwen2.5-VL-7B + ArtiAgent · RichHF localizationmIoU: 0.126 (GPT-5) vs 0.119Hurt
fine-tune on ArtiAgent dataInternVL3.5-8B · RichHF localizationmIoU: 0.013 -> 0.100Improved
increase ArtiAgent training data from 1K to 100K (ablation)Qwen2.5-VL-7B · detection, localization, explanationaverage across benchmarksImproved
replace human SynthScars half by ArtiAgent samples (1K) (ablation)Qwen2.5-VL-7B · ArtiBench detectionaccuracy: 0.555 -> 0.555No clear effect
replace human SynthScars half by ArtiAgent samples (1K) (ablation)Qwen2.5-VL-7B · ArtiBench localizationmIoU: 0.094 -> 0.074Hurt
replace human SynthScars half by ArtiAgent samples (1K) (ablation)Qwen2.5-VL-7B · ArtiBench explanationROUGE: 0.156 -> 0.222Improved
test-time search with ArtiAgent reward modelFLUX-schnell · MS-COCO promptsdelta reward: 0 -> 0.23±0.08Improved
use DiffDoctor trained on human labelsDiffDoctor · LOKI vs ArtiBench localizationmIoU: 0.175 -> 0.081Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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