# VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation

arXiv: https://arxiv.org/abs/2506.03930 · HTML: https://mlsanity.com/paper/2506.03930/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 28 references that are in the map.
- First posted on arXiv 4 Jun 2025.

## Builds on (most cited first, 12 of 12)

- [SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuning](https://mlsanity.com/paper/2408.05517/) (2024)
- [OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement](https://mlsanity.com/paper/2402.14658/) (2024)
- [RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning](https://mlsanity.com/paper/2410.02089/) (2024)
- [CAD-Llama: Leveraging Large Language Models for Computer-Aided Design Parametric 3D Model Generation](https://mlsanity.com/paper/2505.04481/) (2025)
- [Scaling Text-Rich Image Understanding via Code-Guided Synthetic Multimodal Data Generation](https://mlsanity.com/paper/2502.14846/) (2025)
- [LeDex: Training LLMs to Better Self-Debug and Explain Code](https://mlsanity.com/paper/2405.18649/) (2024)
- [WaitGPT: Monitoring and Steering Conversational LLM Agent in Data Analysis with On-the-Fly Code Visualization](https://mlsanity.com/paper/2408.01703/) (2024)
- [Multi-Turn Code Generation Through Single-Step Rewards](https://mlsanity.com/paper/2502.20380/) (2025)
- [Drawing Pandas: A Benchmark for LLMs in Generating Plotting Code](https://mlsanity.com/paper/2412.02764/) (2024)
- [Revisit Self-Debugging with Self-Generated Tests for Code Generation](https://mlsanity.com/paper/2501.12793/) (2025)
- [Vi(E)va LLM! A Conceptual Stack for Evaluating and Interpreting Generative AI-based Visualizations](https://mlsanity.com/paper/2402.02167/) (2024)
- [The Art of Repair: Optimizing Iterative Program Repair with Instruction-Tuned Models](https://mlsanity.com/paper/2505.02931/) (2025)

## Built on by (most cited first, 17 of 17)

- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [AI for Auto-Research: Roadmap & User Guide](https://mlsanity.com/paper/2605.18661/) (2026)
- [JanusCoder: Towards a Foundational Visual-Programmatic Interface for Code Intelligence](https://mlsanity.com/paper/2510.23538/) (2025)
- [Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility](https://mlsanity.com/paper/2601.17027/) (2026)
- [VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction](https://mlsanity.com/paper/2602.13294/) (2026)
- [Code Aesthetics with Agentic Reward Feedback](https://mlsanity.com/paper/2510.23272/) (2025)
- [VisCoder2: Building Multi-Language Visualization Coding Agents](https://mlsanity.com/paper/2510.23642/) (2025)
- [DiscoverLLM: From Executing Intents to Discovering Them](https://mlsanity.com/paper/2602.03429/) (2026)
- [Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence](https://mlsanity.com/paper/2606.15932/) (2026)
- [InteractScience: Programmatic and Visually-Grounded Evaluation of Interactive Scientific Demonstration Code Generation](https://mlsanity.com/paper/2510.09724/) (2025)
- [RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation](https://mlsanity.com/paper/2603.25804/) (2026)
- [Rubrics as Visual-Repair Context for Self-Evolving UI-to-Code Generation](https://mlsanity.com/paper/2608.24138/) (2026)
- [DV-World: Benchmarking Data Visualization Agents in Real-World Scenarios](https://mlsanity.com/paper/2604.25914/) (2026)
- [Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards](https://mlsanity.com/paper/2607.04727/) (2026)
- [EdiTikZ: Scientific Figure Editing from Revision Trajectories](https://mlsanity.com/paper/2609.01409/) (2026)
- [Leum-VL Technical Report](https://mlsanity.com/paper/2603.20354/) (2026)
- [PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis](https://mlsanity.com/paper/2610.20538/) (2026)

## Abstract (arXiv)

Large language models (LLMs) often struggle with visualization tasks like plotting diagrams, charts, where success depends on both code correctness and visual semantics. Existing instruction-tuning datasets lack execution-grounded supervision and offer limited support for iterative code correction, resulting in fragile and unreliable plot generation. We present VisCode-200K, a large-scale instruction tuning dataset for Python-based visualization and self-correction. It contains over 200K examples from two sources: (1) validated plotting code from open-source repositories, paired with natural language instructions and rendered plots; and (2) 45K multi-turn correction dialogues from Code-Feedback, enabling models to revise faulty code using runtime feedback. We fine-tune Qwen2.5-Coder-Instruct on VisCode-200K to create VisCoder, and evaluate it on PandasPlotBench. VisCoder significantly outperforms strong open-source baselines and approaches the performance of proprietary models like GPT-4o-mini. We further adopt a self-debug evaluation protocol to assess iterative repair, demonstrating the benefits of feedback-driven learning for executable, visually accurate code generation.