Luminate

Structured Generation and Exploration of Design Space with Large Language Models
``If you want to have good ideas, you must have lots of ideas and learn to throw away the bad ones.'' - Linus Pauling, a two-time Nobel prize winner

Abstract

Thanks to their generative capabilities, large language models (LLMs) have become an invaluable tool for creative processes. These models have the capacity to produce hundreds and thousands of visual and textual outputs, offering abundant inspiration for creative endeavors. However, the challenge lies in harnessing this creative potential through structured generation and exploration of these outputs. To address this challenge, we propose a framework that facilitates the structured generation of design space in which users can seamlessly explore, evaluate, and synthesize a multitude of responses. We demonstrate the feasibility and usability of this framework through the design and development of Luminate and user study with 8 professional writers. Our work advances the way users interact with LLMs for creative tasks, empowering them to harness the creative potential of LLMs through the structured generation and exploration of outputs in the creative process.

lumiante interface
Figure 1. Luminate employs structured multi-output approach to enable structured generation and exploration of design space using large language models.
lumiante interface
Figure 2. Luminate employs structured multi-output approach to enable structured generation and exploration of design space using large language models.
lumiante interface
Figure 2. Luminate employs structured multi-output approach to enable structured generation and exploration of design space using large language models.

Publication

Sangho Suh, Meng Chen, Bryan Min, Haijun Xia and Toby Li. 2023. Structured Generation and Exploration of Design Space with Large Language Models.

Cite this work

@article{suh2023luminate, title = {Structured Generation and Exploration of Design Space with Large Language Models for Human-AI Co-Creation}, author = {Suh, Sangho and Chen, Meng and Min, Bryan and Li, Toby Jia-Jun and Xia, Haijun}, journal = {arXiv preprint arXiv:2310.12953}, year = {2023}, }