FigMirror is a new framework that converts scientific figures into executable code, enabling the creation of new plots that preserve the visual style of a reference figure. It achieves this by using Grounded Measurement to locate visual elements via coordinates and measure their properties through code, overcoming limitations of pixel‑level optimization used by prior methods. The authors also introduce PlotTwin‑Bench, a benchmark with fine‑grained code and image‑level style metrics, and demonstrate that FigMirror consistently outperforms existing style‑transfer approaches.
arXiv:2601. 04390v2 Announce Type: replace Abstract: High-quality methodology figures are central to scientific communication, yet they remain difficult and time-consuming to create.
arXiv:2608. 12611v1 Announce Type: cross Abstract: Existing screenshot-to-code systems face a trade-off between flexibility and controllability.
The paper introduces “FigTree”, a multi-agent system that automatically converts a scientific paper into a structured vector figure by recursively constructing SVG programs. It decomposes figures into hierarchical regions, generates each region as a short SVG program, and assembles them, using a render‑critic refinement loop to trace and repair visual defects. Evaluations show that “FigTree” produces high‑quality figures and allows more effective editing than raster‑based methods.
Scientific methodology figures are essential for communicating complex methods clearly, yet creating them remains labor-intensive and typically requires multiple rounds of refinement. Recent image-gen...
arXiv:2509. 05208v2 Announce Type: replace-cross Abstract: Large language models (LLMs) excel at program synthesis, yet their ability to produce symbolic graphics programs (SGPs) that render into precise visual content remains underexplored.
Vision-language models (VLMs) have shown strong capabilities in generating visualization code from textual or visual specifications. However, real-world visualization authoring is inherently iterative: users frequently revise existing visualizations to repair flawed charts or adapt them to desired styles.
arXiv:2608. 03270v1 Announce Type: cross Abstract: GUI grounding maps natural-language instructions to click locations and is essential for reliable GUI agents.
arXiv:2604. 14262v2 Announce Type: replace-cross Abstract: GUI grounding models report over 85% accuracy on standard benchmarks, yet drop 27-56 percentage points when instructions require spatial reasoning rather than direct element naming.
arXiv:2607. 15272v1 Announce Type: cross Abstract: Editing the figures in a research paper is a routine and time-consuming part of everyday research practice: authors relabel components, rearrange panels, and restyle visuals as they revise their manuscripts.
arXiv:2607. 18116v1 Announce Type: new Abstract: Recent work leverages Large Language Models (LLMs) to generate executable code for pedagogical animations using libraries such as Manim.
The paper surveys Multimodal Code Intelligence, focusing on tasks where code is generated, edited, refined, or reasoned about under visually grounded inputs such as screenshots, charts, and videos. It categorizes the field by the role of code—rendered artifact, editable structure, intermediate reasoning trace, or executable tool interface—and organizes benchmarks into four domains: Graphical User Interface, Scientific Visualization, Structured Graphics, and Frontier Tasks and Frameworks. The authors argue that reliable evaluation must include evidence of semantics and interaction beyond visual fidelity, and propose four verification-centered research directions to advance the field toward evidence-grounded executable systems.
Despite rapid advances in generative models, achieving pixel-level precision in sketch-based image editing remains a persistent challenge, particularly for fine-grained local deformations. This gap stems primarily from the critical shortage of high-quality, publicly available benchmark datasets that jointly provide geometric constraints and semantic instructions.