TextLap: Customizing Language Models for Text-to-Layout Planning

Jian Chen, Ruiyi Zhang, Yufan Zhou, Jennifer Healey, Jiuxiang Gu +2 more
10/9/2024
cs.CLcs.LG

No official code found. Best implementation: puar-playground/TextLap (18 stars, MIT, last commit 2024-10-21).

Abstract

Automatic generation of graphical layouts is crucial for many real-world applications, including designing posters, flyers, advertisements, and graphical user interfaces. Given the incredible ability of Large language models (LLMs) in both natural language understanding and generation, we believe that we could customize an LLM to help people create compelling graphical layouts starting with only text instructions from the user. We call our method TextLap (text-based layout planning). It uses a curated instruction-based layout planning dataset (InsLap) to customize LLMs as a graphic designer. We demonstrate the effectiveness of TextLap and show that it outperforms strong baselines, including GPT-4 based methods, for image generation and graphical design benchmarks.

View on arXivView PDF

Code Implementations(1)

puar-playground/TextLapCommunity100%
MIT

Source code of the TextLap model, a LLM for text-2-layout generation.

180Shell, PythonOct 21, 20242 years agoMIT

Similar Papers

VASCAR: Content-Aware Layout Generation via Visual-Aware Self-Correction

85% match2024

ChatPattern: Layout Pattern Customization via Natural Language

84% match2024

LayoutPrompter: Awaken the Design Ability of Large Language Models

84% match2 repos2102026

Large Language Models Understand Layout

83% match1 repo72024

From Pixels to Policies: Reinforcing Spatial Reasoning in Language Models for Content-Aware Layout Design

82% match2026

Automatic Layout Planning for Visually-Rich Documents with Instruction-Following Models

82% match2024

Cite this paper

@article{chen2024textlap,
  title  = {TextLap: Customizing Language Models for Text-to-Layout Planning},
  author = {Jian Chen and Ruiyi Zhang and Yufan Zhou and Jennifer Healey and Jiuxiang Gu and Zhiqiang Xu and Changyou Chen},
  year   = {2024},
  eprint = {2410.12844},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2410.12844v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2410.12844.svg)](https://codeofpaper.com/paper/2410.12844)

Discussion