AI Paper Atlas

Research map / Image generation / Editing and style transfer

CHARTSTYLE-100K: A Large-Scale Dataset for Structured Visualization Style Transfer

Yuwei Yang, Tianchi Xie, Jinhong Ni, Yukai Guo, Jing Zhang, Liang Zheng, Yalong Bai, YUHUI YUAN

ECCV 2026 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer

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 general-domain style transfer fine-tuning (stage 1) (ablation)Qwen-Image-Edit · ChartStyle-BenchContent leakage: 0.51 -> 0.13Improved
add general-domain style transfer fine-tuning (stage 1) (ablation)Qwen-Image-Edit · ChartStyle-BenchOverall: 1.58 -> 2.45Improved
add chart-only fine-tuning (stage 2) without stage 1 (ablation)Qwen-Image-Edit · ChartStyle-BenchContent consistency: 2.75 -> 2.45Hurt
add chart-only fine-tuning (stage 2) (ablation)Qwen-Image-Edit · ChartStyle-BenchContent leakage: 0.51 -> 0.03Improved
add stage 1 before stage 2 (progressive training) (ablation)Qwen-Image-Edit · ChartStyle-BenchOverall: 2.53 -> 3.03Improved
add stage 1 before stage 2 (progressive training) (ablation)Qwen-Image-Edit · ChartStyle-BenchOCRScore: 0.54 -> 0.76Improved
add style-space resampling (ablation)ReChart · ChartStyle-BenchOverall: 2.71 -> 3.03Improved
add style-space resampling (ablation)ReChart · ChartStyle-BenchStyle similarity: 2.75 -> 2.90Improved
add style-space resampling (ablation)ReChart · ChartStyle-BenchContent consistency: 3.57 -> 3.96Improved
larger text sizeReChart · ChartStyle-Bench (text size buckets)OCR F1: 0.56 -> 0.78Improved
ReChart vs GPT-Image-1.5ReChart · ChartStyle-BenchOverall: 2.82 -> 3.03Improved
pixel-based editing instead of code-based pipelineReChart · ChartStyle-BenchVisual qualityImproved

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