Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Luminate: Structured Generation and Exploration of Design Space with Large Language Models for Human-AI Co-Creation

Sangho Suh, Meng Chen, Bryan Min, Toby Jia-Jun Li, Haijun Xia

TL;DR

The paper tackles the problem of underutilization of LLM creativity due to unstructured interaction that biases users toward narrow results. It proposes a design-space thinking framework, Prompting for Design Space, and implements Luminate, a system that first generates explicit design dimensions from prompts and then produces dimension-guided outputs to enable structured exploration. Through a user study with 14 professional writers, it demonstrates that dimension-driven generation fosters divergent thinking, helps users understand the design space, and offers practical benefits over conventional AI tools, while also highlighting cognitive load considerations. The work contributes a novel framework, a functional prototype, and empirical evidence that explicit design-space exploration can enhance human-AI co-creation, with potential applicability across domains and media beyond text.

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. But are we harnessing their full potential? We argue that current interaction paradigms fall short, guiding users towards rapid convergence on a limited set of ideas, rather than empowering them to explore the vast latent design space in generative models. To address this limitation, 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 usefulness of this framework through the design and development of an interactive system, Luminate, and a user study with 14 professional writers. Our work advances how we interact with LLMs for creative tasks, introducing a way to harness the creative potential of LLMs.

Luminate: Structured Generation and Exploration of Design Space with Large Language Models for Human-AI Co-Creation

TL;DR

The paper tackles the problem of underutilization of LLM creativity due to unstructured interaction that biases users toward narrow results. It proposes a design-space thinking framework, Prompting for Design Space, and implements Luminate, a system that first generates explicit design dimensions from prompts and then produces dimension-guided outputs to enable structured exploration. Through a user study with 14 professional writers, it demonstrates that dimension-driven generation fosters divergent thinking, helps users understand the design space, and offers practical benefits over conventional AI tools, while also highlighting cognitive load considerations. The work contributes a novel framework, a functional prototype, and empirical evidence that explicit design-space exploration can enhance human-AI co-creation, with potential applicability across domains and media beyond text.

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. But are we harnessing their full potential? We argue that current interaction paradigms fall short, guiding users towards rapid convergence on a limited set of ideas, rather than empowering them to explore the vast latent design space in generative models. To address this limitation, 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 usefulness of this framework through the design and development of an interactive system, Luminate, and a user study with 14 professional writers. Our work advances how we interact with LLMs for creative tasks, introducing a way to harness the creative potential of LLMs.
Paper Structure (44 sections, 11 figures, 1 table)

This paper contains 44 sections, 11 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Supporting Creative Process
  4. Co-Creating with AI
  5. Facilitating Exploration and Sensemaking
  6. Prompting for Design Space: A Framework for Design Space Thinking in Human-AI Co-Creation
  7. Current Interaction Paradigms (Figs. \ref{['fig:teaser']}A & B)
  8. Structured Multi-Output Approach (Fig. \ref{['fig:teaser']}C)
  9. Motivation
  10. Design Goals
  11. Luminate
  12. Interface & Features
  13. Dimension Generation (Fig. \ref{['fig:pipeline']}, DG1)
  14. Dimension-Guided Response Generation (Fig. \ref{['fig:pipeline']}, DG2)
  15. Dimension Selection (Fig. \ref{['fig:dimension-select']}, DG5)
  16. ...and 29 more sections

Figures (11)

  • Figure 1: An example showing the benefits of exposing design space to users. By generating relevant dimensions for a task or topic, we can inform users about many dimensions and [values] they can consider --- e.g., plot complexity: [linear, twisting, ...], genre: [fantasy, adventure, ...]. If all the possible dimensions and values are used to generate responses, the responses will cover many sub-spaces within the overall design space and reveal to users a space of possible responses they can generate with LLMs.
  • Figure 2: Luminate interface consists of two main sections: (A) text editor and (F) exploration view. Users can (B) type and use various (C) text styles (e.g., Title, normal text). They have the option to (D) input a prompt (e.g., ) to LLM and view (E) one of the generated responses in the text editor. (F) In the exploration view, users can observe (J) a cluster of responses and (H) dimensions. At first, no dimension is selected, to invite users to explore the design space. Users can add dimensions to the axes (I) and arrange responses by their dimension values (see Fig. \ref{['fig:dimension-select']}). For optimal experience, we promptly display (G) dimensions as soon as they are generated, allowing users to examine them while responses are being generated. Users can click (K) semantic level icons to adjust the zoom scale and view responses at different levels of detail (see Fig. \ref{['fig:semantic-zoom']}).
  • Figure 3: Technical pipeline of Prompting for Design Space: The pipeline consists of two LLM prompting steps. (A) The first LLM step feeds the inputs from the text editor into the LLM API call to generate dimensions and their values as a JSON object. (B) Luminate transforms the generated object to form a list of requirements per response by randomly selecting a value from each dimension. The second LLM step feeds each list into separate LLM API calls and generates all responses in parallel. (C) Generated responses are then visualized in Luminate's response space.
  • Figure 4: Dimension selection: upon (1) selecting a dimension (Mood) for the x-axis, responses (2) reposition to vertically align to their dimension values (Romantic, Somber, Cheerful, Vengeful). (3) Selecting a dimension for the y-axis (Tone) additionally repositions the responses to align horizontally to their dimension values (Motivating, Hopeful, Inspiring, Peaceful).
  • Figure 5: Two methods for generating similar responses: Method#1: (A) Users can click the button on the node to generate similar responses. Method#2: (B) When nodes are filtered on the canvas, the user can click the button to generate more nodes to be added to the filtered subspace; (C) newly generated nodes join the canvas and align themselves to their dimension values.
  • ...and 6 more figures