# Edit2TikZ: A Comprehensive and Challenging Benchmark for Scientific Figure Editing with TikZ

arXiv: https://arxiv.org/abs/2608.13441 · HTML: https://mlsanity.com/paper/2608.13441/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 15 of its 20 references that are in the map.
- First posted on arXiv 13 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Chart understanding and anomaly detection](https://mlsanity.com/mllm/chart-understanding-and-anomaly-detection/)

## Builds on (most cited first, 15 of 15)

- [Design2Code: Benchmarking Multimodal Code Generation for Automated Front-End Engineering](https://mlsanity.com/paper/2403.03163/) (2024)
- [ChartMimic: Evaluating LMM's Cross-Modal Reasoning Capability via Chart-to-Code Generation](https://mlsanity.com/paper/2406.09961/) (2024)
- [ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation](https://mlsanity.com/paper/2501.06598/) (2025)
- [ChartEdit: How Far Are MLLMs From Automating Chart Analysis? Evaluating MLLMs' Capability via Chart Editing](https://mlsanity.com/paper/2505.11935/) (2025)
- [TikZero: Zero-Shot Text-Guided Graphics Program Synthesis](https://mlsanity.com/paper/2503.11509/) (2025)
- [SVGEditBench V2: A Benchmark for Instruction-based SVG Editing](https://mlsanity.com/paper/2502.19453/) (2025)
- [ChartEditor: A Reinforcement Learning Framework for Robust Chart Editing](https://mlsanity.com/paper/2511.15266/) (2025)
- [ChartM3: Benchmarking Chart Editing with Multimodal Instructions](https://mlsanity.com/paper/2507.21167/) (2025)
- [VCG-Bench: Towards A Unified Visual-Centric Benchmark for Structured Generation and Editing](https://mlsanity.com/paper/2605.15677/) (2026)
- [ChartEditBench: Evaluating Grounded Multi-Turn Chart Editing in Multimodal Language Models](https://mlsanity.com/paper/2602.15758/) (2026)
- [Scientific Graphics Program Synthesis via Dual Self-Consistency Reinforcement Learning](https://mlsanity.com/paper/2604.06079/) (2026)
- [LLM Code Customization with Visual Results: A Benchmark on TikZ](https://mlsanity.com/paper/2505.04670/) (2025)
- [Vision2Code: A Multi-Domain Benchmark for Evaluating Image-to-Code Generation](https://mlsanity.com/paper/2605.11307/) (2026)
- [Evaluating Image Editing with LLMs: A Comprehensive Benchmark and Intermediate-Layer Probing Approach](https://mlsanity.com/paper/2603.19775/) (2026)
- [GeoTikzBridge: Advancing Multimodal Code Generation for Geometric Perception and Reasoning](https://mlsanity.com/paper/2603.22687/) (2026)

## Built on by (most cited first, 1 of 1)

- [Pixels or Code? Benchmarking Instruction-Guided Visual Artifact Operation](https://mlsanity.com/paper/2610.16444/) (2026)

## Abstract (arXiv)

Although multimodal large language models (MLLMs) have shown substantial potential in visual understanding and graphic code generation, editing scientific figures through code presents a greater challenge: a model must jointly recover visual structure, ground the requested change, generate compilable code, and preserve all unrelated content. While existing TikZ benchmarks mainly focus on figure reconstruction and generation, few systematically evaluate instruction-guided scientific figure editing with compilable code. We introduce Edit2TikZ, a comprehensive benchmark for scientific figure editing tasks, featuring 1,548 diverse and high-quality samples. Edit2TikZ combines real-world and controlled synthetic edit cases, supports both textual and visual localization request, and contains multi-step editing, each with step-level annotations. We further construct a human-aligned evaluation framework to measure whether a requested edit is completed while irrelevant content is preserved. Utilizing Edit2TikZ, we evaluate 14 mainstream MLLMs and find that current systems remain unreliable: on average, proprietary models achieve a compilation success rate of merely 75% and remain limited in both figure restoration and edit correctness, while compact models below 9B struggle further with instruction following and complete figure generation. Therefore, we build a mixed training set TikZEditMix and adopt reconstruction-then-editing curriculum learning for compact models. On Qwen3.5-4B, this training improves the compilation success rate from 45.35% to 83.40% and yields an average improvement of 18.7 points across our proposed evaluation metrics. The code and data will be released at https://github.com/Solunny/Edit2TikZ.