Research map / Image generation / Autoregressive and latent generation
Charts Are Not Images: On the Challenges of Scientific Chart Editing
Li Li, Ryan A. Rossi, Sungchul Kim, Sunav Choudhary, Franck Dernoncourt, Puneet Mathur, Zhengzhong Tu, Yue Zhao
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 changed | Where | Measured | Effect |
|---|---|---|---|
| compare Imagen 4 SSIM against its instruction score | Imagen 4 · FigEdit Single edit | SSIM vs instruction score: SSIM 0.7726 vs instruction 1.58 | Mixed |
| use GPT-Image instead of Imagen 4 for conversational edits | Imagen 4 · FigEdit Conversational edit | instruction-following score: 1.35 -> 4.59 | Improved |
| use GPT-Image instead of Imagen 4 for conversational edits (SSIM) | Imagen 4 · FigEdit Conversational edit | SSIM: 0.7180 -> 0.6732 | Hurt |
| use OmniGen2 instead of InstructPix2Pix on single edits | InstructPix2Pix · FigEdit Single edit | instruction-following score: 3.27 -> 3.35 | No clear effect |
| use OmniGen2 instead of GPT-Image on visual guidance | GPTImage · FigEdit Visual guidance | instruction-following score: 2.39 -> 1.19 | Hurt |
| use OmniGen2 instead of GPT-Image on style transfer | GPTImage · FigEdit Style transfer | instruction-following score: 3.06 -> 1.53 | Hurt |
| go from single to multi edits | Imagen 4 · FigEdit | instruction-following score: 1.58 -> 1.26 | Hurt |
| go from single to multi edits | OmniGen2 · FigEdit | instruction-following score: 3.35 -> 2.65 | Hurt |
| go from single to multi edits | GPTImage · FigEdit | instruction-following score: 3.47 -> 2.51 | Hurt |
| use GPT-Image on style transfer vs single | GPTImage · FigEdit | SSIM and instruction score: SSIM 0.7295 -> 0.8438; instruction 3.47 -> 3.06 | Mixed |
Closest papers by meaning
- ChartE$^{3}$: A Comprehensive Benchmark for End-to-End Chart Editing (ICML 2026 · similarity 0.94)
- ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding (CVPR 2026 · similarity 0.88)
- Factuality Matters: When Image Generation and Editing Meet Structured Visuals (ICLR 2026 · similarity 0.87)
- Visual Self-Refine: A Pixel-Guided Paradigm for Accurate Chart Parsing (ICLR 2026 · similarity 0.86)
- EvoChart: A Benchmark and a Self-Training Approach Towards Real-World Chart Understanding (AAAI 2025 · similarity 0.85)
- ChartGalaxy: A Dataset for Infographic Chart Understanding and Generation (ICLR 2026 · similarity 0.85)
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.
