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

2025-06-04 · Yuansheng Ni, Ping Nie, Kai Zou, Xiang Yue, Wenhu Chen

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.

📄 PDF Abstract BibTeX arXiv:2506.03930

Code (0)

등록된 구현이 없습니다.

Tasks

Code Generation

Methods 이 논문이 사용한 방법론

ADOPT Please enter a description about the method here

VisCoder2: Building Multi-Language Visualization Coding Agents

2025-10-24 · Yuansheng Ni, Songcheng Cai, Xiangchao Chen, Jiarong Liang 외 arxiv

Large language models (LLMs) have recently enabled coding agents capable of generating, executing, and revising visualization code. However, existing models often fail in practical workflows due to limited language cover…

Executable Code Actions Elicit Better LLM Agents

2024-02-01 · Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang 외

Large Language Model (LLM) agents, capable of performing a broad range of actions, such as invoking tools and controlling robots, show great potential in tackling real-world challenges. LLM agents are typically prompted …

Language Modelling Large Language Model

BRIDGE: Building Representations In Domain Guided Program Synthesis

2025-11-26 · Robert Joseph George, Carson Eisenach, Udaya Ghai, Dominique Perrault-Joncas 외 arxiv

Large language models can generate plausible code, but remain brittle for formal verification in proof assistants such as Lean. A central scalability challenge is that verified synthesis requires consistent artifacts acr…

Program Synthesis

CodeSpecBench: Benchmarking LLMs for Executable Behavioral Specification Generation

2026-04-14 · Zaoyu Chen, Jianbo Dai, Boyu Zhu, Jingdong Wang 외 arxiv

Large language models (LLMs) can generate code from natural language, but the extent to which they capture intended program behavior remains unclear. Executable behavioral specifications, defined via preconditions and po…

Code Generation

Exploring different approaches to customize language models for domain-specific text-to-code generation

2026-03-17 · Luís Freire, Fernanda A. Andaló, Nicki Skafte Detlefsen arxiv

Large language models (LLMs) have demonstrated strong capabilities in generating executable code from natural language descriptions. However, general-purpose models often struggle in specialized programming contexts wher…

parameter-efficient fine-tuning Text-to-Code Generation Few-Shot Learning