Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ChatPattern: Layout Pattern Customization via Natural Language

Zixiao Wang, Yunheng Shen, Xufeng Yao, Wenqian Zhao, Yang Bai, Farzan Farnia, Bei Yu

TL;DR

Experiments on challenging pattern generation setting shows the ability of ChatPattern to synthesize high-quality large-scale patterns.

Abstract

Existing works focus on fixed-size layout pattern generation, while the more practical free-size pattern generation receives limited attention. In this paper, we propose ChatPattern, a novel Large-Language-Model (LLM) powered framework for flexible pattern customization. ChatPattern utilizes a two-part system featuring an expert LLM agent and a highly controllable layout pattern generator. The LLM agent can interpret natural language requirements and operate design tools to meet specified needs, while the generator excels in conditional layout generation, pattern modification, and memory-friendly patterns extension. Experiments on challenging pattern generation setting shows the ability of ChatPattern to synthesize high-quality large-scale patterns.

ChatPattern: Layout Pattern Customization via Natural Language

TL;DR

Experiments on challenging pattern generation setting shows the ability of ChatPattern to synthesize high-quality large-scale patterns.

Abstract

Existing works focus on fixed-size layout pattern generation, while the more practical free-size pattern generation receives limited attention. In this paper, we propose ChatPattern, a novel Large-Language-Model (LLM) powered framework for flexible pattern customization. ChatPattern utilizes a two-part system featuring an expert LLM agent and a highly controllable layout pattern generator. The LLM agent can interpret natural language requirements and operate design tools to meet specified needs, while the generator excels in conditional layout generation, pattern modification, and memory-friendly patterns extension. Experiments on challenging pattern generation setting shows the ability of ChatPattern to synthesize high-quality large-scale patterns.
Paper Structure (12 sections, 11 equations, 10 figures, 1 table)

This paper contains 12 sections, 11 equations, 10 figures, 1 table.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. The Scope of $\mathsf{ChatPattern}$
  4. Fixed-size Layout Pattern Generation
  5. Free-size Layout Pattern Generation
  6. Algorithm
  7. Pattern Customization via Expert LLM agent
  8. Flexible Layout Pattern Generative Model
  9. Experimental Results
  10. Layout Pattern Generation
  11. Evaluation of LLM agent
  12. Conclusion

Figures (10)

  • Figure 1: An illustration of ChatPattern.
  • Figure 2: Squish Pattern Representation.
  • Figure 3: Design rule illustrations. 'Space' means distance between adjacent polygons. 'Width' measures shape size in one direction. 'Area' denotes area of a polygon.
  • Figure 4: A working pipeline of LLM agent in $\mathsf{ChatPattern}$. Dash lines denote the optional paths.
  • Figure 5: Illustration of conditional pattern generation.
  • ...and 5 more figures

Theorems & Definitions (2)

  • Definition 1: Legality
  • Definition 2: Diversity