# ChartEditBench: Evaluating Grounded Multi-Turn Chart Editing in Multimodal Language Models

arXiv: https://arxiv.org/abs/2602.15758 · HTML: https://mlsanity.com/paper/2602.15758/

## Key facts

- Cited by 8 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 11 of its 24 references that are in the map.
- First posted on arXiv 17 Feb 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, 11 of 11)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [ViperGPT: Visual Inference via Python Execution for Reasoning](https://mlsanity.com/paper/2303.08128/) (2023)
- [PlotQA: Reasoning over Scientific Plots](https://mlsanity.com/paper/1909.00997/) (2019)
- [FigureQA: An Annotated Figure Dataset for Visual Reasoning](https://mlsanity.com/paper/1710.07300/) (2017)
- [Can You Trust LLM Judgments? Reliability of LLM-as-a-Judge](https://mlsanity.com/paper/2412.12509/) (2024)
- [Improving LLM-as-a-Judge Inference with the Judgment Distribution](https://mlsanity.com/paper/2503.03064/) (2025)
- [How Reliable is Multilingual LLM-as-a-Judge?](https://mlsanity.com/paper/2505.12201/) (2025)
- [Are We on the Right Way to Assessing LLM-as-a-Judge?](https://mlsanity.com/paper/2512.16041/) (2025)
- [From Charts to Code: A Hierarchical Benchmark for Multimodal Models](https://mlsanity.com/paper/2510.17932/) (2025)
- [Exploring How Generative MLLMs Perceive More Than CLIP with the Same Vision Encoder](https://mlsanity.com/paper/2411.05195/) (2024)
- [Vision-Language Models as Differentiable Semantic and Spatial Rewards for Text-to-3D Generation](https://mlsanity.com/paper/2509.15772/) (2025)

## Built on by (most cited first, 8 of 8)

- [LLMs Corrupt Your Documents When You Delegate](https://mlsanity.com/paper/2604.15597/) (2026)
- [Edit2TikZ: A Comprehensive and Challenging Benchmark for Scientific Figure Editing with TikZ](https://mlsanity.com/paper/2608.13441/) (2026)
- [ChartAnno: Benchmarking Multimodal Large Language Models for Chart Annotation Generation](https://mlsanity.com/paper/2608.03464/) (2026)
- [Charts Are Beyond Pixels: Probing for Layer-Wise Chart Understanding and Editing](https://mlsanity.com/paper/2609.08657/) (2026)
- [Point, Revise, Review: Grounded Agentic Analysis in Reactive Notebooks with marimo-lens](https://mlsanity.com/paper/2609.19839/) (2026)
- [ChartRevise: A Dataset and Evaluation Protocol for Exact Chart Editing via Code](https://mlsanity.com/paper/2609.38642/) (2026)
- [VisAudit: Evaluating Multimodal Agents for Visual Diagnosis and Repair](https://mlsanity.com/paper/2610.02399/) (2026)
- [Pixels or Code? Benchmarking Instruction-Guided Visual Artifact Operation](https://mlsanity.com/paper/2610.16444/) (2026)

## Abstract (arXiv)

While Multimodal Large Language Models (MLLMs) perform strongly on single-turn chart generation, their ability to support real-world exploratory data analysis remains underexplored. In practice, users iteratively refine visualizations through multi-turn interactions that require maintaining common ground, tracking prior edits, and adapting to evolving preferences. We introduce ChartEditBench, a benchmark for incremental, visually grounded chart editing via code, comprising 5,000 difficulty-controlled modification chains and a rigorously human-verified subset. Unlike prior one-shot benchmarks, ChartEditBench evaluates sustained, context-aware editing. We further propose a robust evaluation framework that mitigates limitations of LLM-as-a-Judge metrics by integrating execution-based fidelity checks, pixel-level visual similarity, and logical code verification. Experiments with state-of-the-art MLLMs reveal substantial degradation in multi-turn settings due to error accumulation and breakdowns in shared context, with strong performance on stylistic edits but frequent execution failures on data-centric transformations. ChartEditBench, establishes a challenging testbed for grounded, intent-aware multimodal programming.