Seeing Twice: How Side-by-Side T2I Comparison Changes Auditing Strategies
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915639404265472 |
|---|---|
| author | Maldaner, Matheus Kunzler Deng, Wesley Hanwen Hong, Jason I. Holstein, Kenneth Eslami, Motahhare |
| author_facet | Maldaner, Matheus Kunzler Deng, Wesley Hanwen Hong, Jason I. Holstein, Kenneth Eslami, Motahhare |
| contents | While generative AI systems have gained popularity in diverse applications, their potential to produce harmful outputs limits their trustworthiness and utility. A small but growing line of research has explored tools and processes to better engage non-AI expert users in auditing generative AI systems. In this work, we present the design and evaluation of MIRAGE, a web-based tool exploring a "contrast-first" workflow that allows users to pick up to four different text-to-image (T2I) models, view their images side-by-side, and provide feedback on model performance on a single screen. In our user study with fifteen participants, we used four predefined models for consistency, with only a single model initially being shown. We found that most participants shifted from analyzing individual images to general model output patterns once the side-by-side step appeared with all four models; several participants coined persistent "model personalities" (e.g., cartoonish, saturated) that helped them form expectations about how each model would behave on future prompts. Bilingual participants also surfaced a language-fidelity gap, as English prompts produced more accurate images than Portuguese or Chinese, an issue often overlooked when dealing with a single model. These findings suggest that simple comparative interfaces can accelerate bias discovery and reshape how people think about generative models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_21547 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Seeing Twice: How Side-by-Side T2I Comparison Changes Auditing Strategies Maldaner, Matheus Kunzler Deng, Wesley Hanwen Hong, Jason I. Holstein, Kenneth Eslami, Motahhare Human-Computer Interaction While generative AI systems have gained popularity in diverse applications, their potential to produce harmful outputs limits their trustworthiness and utility. A small but growing line of research has explored tools and processes to better engage non-AI expert users in auditing generative AI systems. In this work, we present the design and evaluation of MIRAGE, a web-based tool exploring a "contrast-first" workflow that allows users to pick up to four different text-to-image (T2I) models, view their images side-by-side, and provide feedback on model performance on a single screen. In our user study with fifteen participants, we used four predefined models for consistency, with only a single model initially being shown. We found that most participants shifted from analyzing individual images to general model output patterns once the side-by-side step appeared with all four models; several participants coined persistent "model personalities" (e.g., cartoonish, saturated) that helped them form expectations about how each model would behave on future prompts. Bilingual participants also surfaced a language-fidelity gap, as English prompts produced more accurate images than Portuguese or Chinese, an issue often overlooked when dealing with a single model. These findings suggest that simple comparative interfaces can accelerate bias discovery and reshape how people think about generative models. |
| title | Seeing Twice: How Side-by-Side T2I Comparison Changes Auditing Strategies |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2511.21547 |