BROWSE
Detail
Conference Paper
What if I Were?: Synthetic Lived Experience based on Generative Models
- Authors
- Sun-Kyung Lee, Hyuntae Jeong, Se Won Oh, Jeong Mook Lim, Kyoung Ju Noh, Seungeun Chung, Gyuwon Jung
- Issue Date
- 2026-04
- Citation
- Conference on Human Factors in Computing Systems (CHI) 2026, pp.1-5
- ISSN
- 4007-2281
- Publisher
- ACM
- Language
- English
- Type
- Conference Paper
- Abstract
- People often wonder what it would be like to be in a different situation: how others live, and how their own life might unfold under different circumstances. In this work, we propose Synthetic Lived Experience (SLE) as computationally generated, first-person accounts that approximate “what I might do and feel if I were this person, in this situation” or “how my life might look if my context changed.” Using a generative model-based pipeline, we specify personas and contexts, and then synthesize episodic narratives and corresponding visualizations that depict actions, environments, and emotions. We present initial qualitative examples of SLEs that illustrate how generated first-person narratives can surface everyday lived experience under specified personas and contexts. We also discuss key limitations and outline opportunities for future work.
- Keyword
- Large Language Models, Generative Models, Lived Experience, Synthetic, Experience Augmentation
- KSP Keywords
- First-person, Generative models, Lived experience, WHAT If, language models, model-based
This work is distributed under the term of Creative Commons License (CCL)
(CC BY)
(CC BY)
