I Can Embrace and Avoid Vagueness Myself: Supporting the Design Process by Balancing Vagueness through Text-to-Image Generative AI

Myungjin Kim,Bogoan Kim,Kyungsik Han

2024-11-13

Abstract:This study examines the role of vagueness in the design process and its strategic management for the effective human-AI interaction. While vagueness in the generation of design ideas promotes diverse interpretations and prevents fixation, excessive vagueness can lead to scattered results. Designers attempt to use image search tools or generative AIs (e.g., Dall-E) for their work but often fail to achieve satisfactory results because the level of vagueness is not properly managed in these technologies. In this work, we identified how designers coordinate vagueness in their design process and applied key components of the process to the design of CLAY, an interactive system that balances vagueness through iterative prompt refinement by integrating the strengths of text-to-image generative AI. Results from our user study with 10 fashion designers showed that CLAY effectively supported their design process, reducing design time, and expanding creative possibilities compared to their existing practice, by allowing them to both embrace and avoid vagueness as needed. Our study highlights the importance of identifying key characteristics of the target user and domain, and exploring ways to incorporate them into the design of an AI-based interactive tool.

Human-Computer Interaction

What problem does this paper attempt to address?

This paper attempts to solve the problem of how to effectively manage vagueness during the design process, especially when using text - to - image generative AI. Specifically, the research focuses on the following points:

1. **Positive and negative effects of vagueness**:
- Vagueness helps to promote diverse interpretations and avoid early fixation, thus stimulating creativity.
- However, excessive vagueness can lead to scattered results, lack of clear direction, and affect the effectiveness of design and decision - making.

2. **Limitations of existing tools**:
- Designers usually work with image search tools or generative AIs (such as DALL - E), but these tools are often unable to effectively manage vagueness, resulting in unsatisfactory results.
- Traditional image search tools (such as Google Image Search, Pinterest) can provide diverse search results, but may produce inaccurate or irrelevant results when dealing with specific queries, increasing the time and complexity of finding meaningful information.
- Text - to - image generative AI requires specific prompts, which may be beyond the capabilities of designers, limiting their creative advantage of exploring multiple possibilities from vagueness.

3. **Proposed solution: CLAY system**:
- CLAY is an interactive system designed to balance vagueness through iterative prompt refinement.
- This system combines the advantages of text - to - image generative AI, helping designers effectively manage vagueness during the design process, being able to embrace vagueness to stimulate creativity and reduce vagueness when necessary to achieve specific design goals.

4. **User study verification**:
- Through experiments with 10 fashion designers, CLAY has been proven to reduce design time, expand creative possibilities, and allow designers to more actively create the required moodboards and final designs.

### Specific problem summary

- **The role of vagueness in design**: Vagueness is inevitable in the early stages of design, providing designers with space for multiple interpretations and creativity. However, excessive vagueness may lead to unclear design directions, so effective management is required.

- **Deficiencies of existing tools**: Existing image search tools and generative AIs have limitations in dealing with vagueness and cannot fully meet the needs of designers.

- **The role of CLAY**: CLAY helps designers balance vagueness during the design process through iterative prompt refinement, retaining the creative advantages of vagueness while avoiding its negative impacts, thereby improving design efficiency and quality.

### Main contributions of the paper

- **Identifying key structures and challenges in vagueness management**: By studying the hierarchical and combinatorial structures used by designers when balancing vagueness, as well as the challenges they encounter when using existing tools.

- **Developing the CLAY system**: Designing and implementing the CLAY system, integrating the vagueness management process into text - to - image generative AI to support designers to design more effectively.

- **User study and design implications**: Verifying the effectiveness of the CLAY system through user study and discussing the future design directions of text - to - image generative AI based on vagueness management.

Through these efforts, this paper not only solves the practical problem of vagueness management in the design process, but also provides valuable insights for future research and applications.