Pith. sign in

REVIEW 5 cited by

AutomaTikZ: Text-Guided Synthesis of Scientific Vector Graphics with TikZ

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 2310.00367 v2 pith:3XBZL55T submitted 2023-09-30 cs.CL cs.CV

Jonas Belouadi , Anne Lauscher , Steffen Eger This is my paper
classification cs.CLcs.CV
keywords graphicstikzfiguresvectorclimalanguagellamamodel
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Generating bitmap graphics from text has gained considerable attention, yet for scientific figures, vector graphics are often preferred. Given that vector graphics are typically encoded using low-level graphics primitives, generating them directly is difficult. To address this, we propose the use of TikZ, a well-known abstract graphics language that can be compiled to vector graphics, as an intermediate representation of scientific figures. TikZ offers human-oriented, high-level commands, thereby facilitating conditional language modeling with any large language model. To this end, we introduce DaTikZ, the first large-scale TikZ dataset consisting of 120k TikZ drawings aligned with captions. We fine-tune LLaMA on DaTikZ, as well as our new model CLiMA, which augments LLaMA with multimodal CLIP embeddings. In both human and automatic evaluation, CLiMA and LLaMA outperform commercial GPT-4 and Claude 2 in terms of similarity to human-created figures, with CLiMA additionally improving text-image alignment. Our detailed analysis shows that all models generalize well and are not susceptible to memorization. GPT-4 and Claude 2, however, tend to generate more simplistic figures compared to both humans and our models. We make our framework, AutomaTikZ, along with model weights and datasets, publicly available.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

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

  1. 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.

  2. Generating Pedagogically Meaningful Visuals for Math Word Problems: A New Benchmark and Analysis of Text-to-Image Models

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Math2Visual converts math word problems into teacher-co-designed formal and intuitive visuals, and its benchmark shows text-to-image models still frequently fail at depicting mathematical relations.

  3. GeoLoom: High-quality Geometric Diagram Generation from Textual Input

    cs.CV 2025-12 conditional novelty 5.0 of 10

    Natural-language geometry descriptions can be autoformalized into a custom geometry language and converted to coordinates by Monte Carlo optimization, yielding usable diagrams in seconds for about 81-85% of test problems.

  4. Exploring Agentic Workflows for Generating High Quality Math Visual Aids

    cs.AI 2026-07 conditional novelty 4.0 of 10

    An exploratory agentic self-improvement loop for TikZ math diagrams yields modest human-rated gains but fails to fix many spatial and coverage errors.

  5. AI4Research: A Survey of Artificial Intelligence for Scientific Research

    cs.CL 2025-07 conditional novelty 4.0 of 10

    A survey that organizes AI-for-research work into five tasks, comprehension, survey, discovery, writing, and peer review, and compiles associated tools and benchmarks.

Pith tools