Pith. sign in

REVIEW 8 cited by

Empowering LLMs to Understand and Generate Complex Vector Graphics

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2412.11102 v3 pith:2C2GU3FG submitted 2024-12-15 cs.CV

Ximing Xing , Juncheng Hu , Guotao Liang , Jing Zhang , Dong Xu , Qian Yu This is my paper · ORCID
classification cs.CV
keywords vectorllmsdatainstructionlanguagegenerategenerationgraphics
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The unprecedented advancements in Large Language Models (LLMs) have profoundly impacted natural language processing but have yet to fully embrace the realm of scalable vector graphics (SVG) generation. While LLMs encode partial knowledge of SVG data from web pages during training, recent findings suggest that semantically ambiguous and tokenized representations within LLMs may result in hallucinations in vector primitive predictions. Additionally, LLM training typically lacks modeling and understanding of the rendering sequence of vector paths, which can lead to occlusion between output vector primitives. In this paper, we present LLM4SVG, an initial yet substantial step toward bridging this gap by enabling LLMs to better understand and generate vector graphics. LLM4SVG facilitates a deeper understanding of SVG components through learnable semantic tokens, which precisely encode these tokens and their corresponding properties to generate semantically aligned SVG outputs. Using a series of learnable semantic tokens, a structured dataset for instruction following is developed to support comprehension and generation across two primary tasks. Our method introduces a modular architecture to existing large language models, integrating semantic tags, vector instruction encoders, fine-tuned commands, and powerful LLMs to tightly combine geometric, appearance, and language information. To overcome the scarcity of SVG-text instruction data, we developed an automated data generation pipeline that collected our SVGX-SFT Dataset, consisting of high-quality human-designed SVGs and 580k SVG instruction following data specifically crafted for LLM training, which facilitated the adoption of the supervised fine-tuning strategy popular in LLM development.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SVGEval: A Vision-Grounded Framework for Perceptual-Quality Benchmarking and Evaluation in Text-to-SVG Generation

    cs.CV 2026-08 conditional novelty 6.0 of 10

    SVGEval benchmarks and trains an explainable multimodal scorer for perceptual quality of text-to-SVG generation, showing a consistent gap on spatial and structural judgments.

  2. GVR-Coder: A Visual-Feedback Framework for Structured SVG Generation in Complex Document and Meeting Scenarios

    cs.LG 2026-07 conditional novelty 6.0 of 10

    A curriculum-plus-RL-plus-repair SVG agent, trained on a new 100K office-diagram dataset, beats strong LLM baselines on layout-correct logical diagrams from long professional text.

  3. Vector-Bench: Can Models Surgically Edit SVG Code?

    cs.AI 2026-07 conditional novelty 6.0 of 10

    Only 2.35% of 1,360 model outputs pass Vector-Bench's three-gate SVG repair-and-preserve reward, and the best endpoint passes 15.0% despite 43.7% mean repair progress.

  4. SVGen: Interpretable Vector Graphics Generation with Large Language Models

    cs.LG 2025-08 conditional novelty 6.0 of 10

    SVGen fine-tunes 3B to 7B LLMs with curriculum learning, chain-of-thought, and GRPO reinforcement to generate SVG icons from text, reporting better in-distribution quality than larger models.

  5. SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    SVGenius benchmarks 22 LLMs on 2,377 SVG tasks across understanding, editing, and generation, finding universal degradation with complexity.

  6. ViewCraft3D: High-Fidelity and View-Consistent 3D Vector Graphics Synthesis

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A two-stage method that fits 3D Bézier curves to a reconstructed mesh and refines them with a 3D diffusion prior, producing view-consistent 3D vector graphics from a single image in about 30 minutes.

  7. LayerTracer: Cognitive-Aligned Layered SVG Synthesis via Diffusion Transformer

    cs.CV 2025-02 conditional novelty 6.0 of 10

    A diffusion transformer trained on SVG construction sequences generates and vectorizes layered SVG graphics, breaking creation into editable steps.

  8. SlideCoder: Layout-aware RAG-enhanced Hierarchical Slide Generation from Design

    cs.CV 2025-06 conditional novelty 5.0 of 10

    SlideCoder converts slide design images to editable python-pptx code and reports large gains over prior baselines on a new difficulty-tiered benchmark.

Pith tools