AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

Simple Test-Time Refinement for Plot-to-Code Generation via Visual-Code Diagnostics

Peiming Guo, Jinpeng Huang, Yangbin Sun, GaoXiong Cao, Ensheng Shi, Yuchi Ma, Yu Zhang, Meishan Zhang, Baotian Hu, Min zhang

NeurIPS 2026 · Large language models › Reasoning and chain of thought · approach: Test-time compute scaling · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a test-time refinement paradigm for plot-to-code generation in which a model generates code first, then iteratively locates and fixes errors using visual-code diagnostics.

Problem

Plot-to-code generation with MLLMs is typically evaluated with one-pass generation or iterative sampling filtered by verifiers, which does not mirror the human generate-then-debug programming workflow.

Approach

Replaces one-pass or verifier-filtered sampling with a test-time refinement loop that mimics human workflow: generate code, locate errors by comparing rendered output with the target plot, and fix them iteratively.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add visual-code diagnostic verifier (RL-trained 2B) to guide patch refinementQwen3-VL-8B generator · ChartMimiccode execution success E: 77.8 -> 90.0Improved
same, visual score LQwen3-VL-8B generator · ChartMimicvisual score L: 66.2 -> 73.4Improved
same, GPT-4o generatorGPT-4o generator · ChartMimicvisual score L: 81.6 -> 87.6Improved
add RL stage after SFT for verifier (ablation)Ours verifier · ChartMimic (Qwen3-VL-8B)execution success E: 86.3 -> 90.0Improved
train verifier on diagnostic data vs untrained 32B verifier (ablation)Qwen3-VL-8B generator · ChartMimicvisual score L: 70.9 (untrained 32B) -> 73.4Improved
add code localization signal to visual error type (ablation)Verifier · ChartMimic (Qwen3-VL-8B)visual score L: 70.7 -> 73.4Improved
add visual error type to code localization (ablation)Verifier · ChartMimic (Qwen3-VL-8B)visual score L: 72.2 -> 73.4Improved
use Jaccard reward instead of exact match (ablation)Verifier RL · ChartMimic (Qwen3-VL-8B)visual score L: 71.9 -> 73.4Improved
use Jaccard reward instead of set F1 (ablation)Verifier RL · ChartMimic (Qwen3-VL-8B)visual score L: 73.4 vs 73.4No clear effect
use diagnostic refinement instead of multi-agent refinement (token cost)GPT-4o generator · Inference tokenstokens: 11264.0 -> 1859.3Improved
apply diagnostic refinement to specialized ChartMaster-7B generatorChartMaster-7B · ChartMimicvisual score L: 79.4 -> 63.4Hurt
Best-of-N samplingQwen3-VL-8B generator · ChartMimicvisual score L: 66.2 -> 66.9Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy