2025
Un-Straightening Generative AI: How Queer Artists Surface and Challenge Model Normativity
Jordan Taylor, Joel Mire, Franchesca Spektor, Alicia DeVrio, Maarten Sap, Haiyi Zhu, Sarah E Fox
FAccT 2025
We examine value tensions between artists and generative AI systems through a 5-week workshop with 13 queer artists.
Un-Straightening Generative AI: How Queer Artists Surface and Challenge Model Normativity
Jordan Taylor, Joel Mire, Franchesca Spektor, Alicia DeVrio, Maarten Sap, Haiyi Zhu, Sarah E Fox
FAccT 2025
We examine value tensions between artists and generative AI systems through a 5-week workshop with 13 queer artists.
Rejected Dialects: Biases Against African American Language in Reward Models
Joel Mire*, Zubin Trivadi Aysola*, Daniel Chechelnitsky, Nicholas Deas, Chrysoula Zerva, Maarten Sap (* equal contribution)
NAACL (Findings) 2025
We introduce a framework for evaluating dialect biases in reward models and conduct a case study showing biases against African American Language texts.
Rejected Dialects: Biases Against African American Language in Reward Models
Joel Mire*, Zubin Trivadi Aysola*, Daniel Chechelnitsky, Nicholas Deas, Chrysoula Zerva, Maarten Sap (* equal contribution)
NAACL (Findings) 2025
We introduce a framework for evaluating dialect biases in reward models and conduct a case study showing biases against African American Language texts.
2024
HEART-felt Narratives: Tracing Empathy and Narrative Style in Personal Stories with LLMs
Jocelyn Shen, Joel Mire, Hae Won Park, Cynthia Breazeal, Maarten Sap
EMNLP 2024
We propose a theoretical framework of narrative style as it relates to empathy, and use LLMs to quantify narrative style for downstream social and behavioral insights.
HEART-felt Narratives: Tracing Empathy and Narrative Style in Personal Stories with LLMs
Jocelyn Shen, Joel Mire, Hae Won Park, Cynthia Breazeal, Maarten Sap
EMNLP 2024
We propose a theoretical framework of narrative style as it relates to empathy, and use LLMs to quantify narrative style for downstream social and behavioral insights.
The Empirical Variability of Narrative Perceptions of Social Media Texts
Joel Mire, Maria Antoniak, Elliott Ash, Andrew Piper, Maarten Sap
EMNLP 2024
We present a dataset and taxonomy of crowd workers' descriptive perceptions of storytelling, analyzing patterns of disagreement among them and across other annotation contexts, including prescriptive labels from researchers and predictions from LLMs.
The Empirical Variability of Narrative Perceptions of Social Media Texts
Joel Mire, Maria Antoniak, Elliott Ash, Andrew Piper, Maarten Sap
EMNLP 2024
We present a dataset and taxonomy of crowd workers' descriptive perceptions of storytelling, analyzing patterns of disagreement among them and across other annotation contexts, including prescriptive labels from researchers and predictions from LLMs.
Where Do People Tell Stories Online? Story Detection Across Online Communities
Maria Antoniak, Joel Mire, Maarten Sap, Elliott Ash, Andrew Piper
ACL 2024
We develop StorySeeker, a toolkit for detecting stories in online conversations, and use it to illuminate distributional characteristics of storytelling across a community-centric social media platform.
Where Do People Tell Stories Online? Story Detection Across Online Communities
Maria Antoniak, Joel Mire, Maarten Sap, Elliott Ash, Andrew Piper
ACL 2024
We develop StorySeeker, a toolkit for detecting stories in online conversations, and use it to illuminate distributional characteristics of storytelling across a community-centric social media platform.
2020
The Perspectival Limits of Computation: Researching the Present in Satin Island
Joel Mire
Intersect: The Stanford Journal of Science, Technology, and Society 2020
Through a case study on Tom McCarthy's novel, Satin Island, I explore the unique status of the novel in relation to the rationales, methods, and consequences of research ethos committed to the idea that computation should supplant other ways of knowing.
The Perspectival Limits of Computation: Researching the Present in Satin Island
Joel Mire
Intersect: The Stanford Journal of Science, Technology, and Society 2020
Through a case study on Tom McCarthy's novel, Satin Island, I explore the unique status of the novel in relation to the rationales, methods, and consequences of research ethos committed to the idea that computation should supplant other ways of knowing.