A Matter of Perspective(s): Contrasting Human and LLM Argumentation in Subjective Decision-Making on Subtle Sexism
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916621511032832 |
|---|---|
| author | Aoyagui, Paula Akemi Stemmler, Kelsey Ferguson, Sharon Kim, Young-ho Kuzminykh, Anastasia |
| author_facet | Aoyagui, Paula Akemi Stemmler, Kelsey Ferguson, Sharon Kim, Young-ho Kuzminykh, Anastasia |
| contents | In subjective decision-making, where decisions are based on contextual interpretation, Large Language Models (LLMs) can be integrated to present users with additional rationales to consider. The diversity of these rationales is mediated by the ability to consider the perspectives of different social actors. However, it remains unclear whether and how models differ in the distribution of perspectives they provide. We compare the perspectives taken by humans and different LLMs when assessing subtle sexism scenarios. We show that these perspectives can be classified within a finite set (perpetrator, victim, decision-maker), consistently present in argumentations produced by humans and LLMs, but in different distributions and combinations, demonstrating differences and similarities with human responses, and between models. We argue for the need to systematically evaluate LLMs' perspective-taking to identify the most suitable models for a given decision-making task. We discuss the implications for model evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_14052 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Matter of Perspective(s): Contrasting Human and LLM Argumentation in Subjective Decision-Making on Subtle Sexism Aoyagui, Paula Akemi Stemmler, Kelsey Ferguson, Sharon Kim, Young-ho Kuzminykh, Anastasia Human-Computer Interaction In subjective decision-making, where decisions are based on contextual interpretation, Large Language Models (LLMs) can be integrated to present users with additional rationales to consider. The diversity of these rationales is mediated by the ability to consider the perspectives of different social actors. However, it remains unclear whether and how models differ in the distribution of perspectives they provide. We compare the perspectives taken by humans and different LLMs when assessing subtle sexism scenarios. We show that these perspectives can be classified within a finite set (perpetrator, victim, decision-maker), consistently present in argumentations produced by humans and LLMs, but in different distributions and combinations, demonstrating differences and similarities with human responses, and between models. We argue for the need to systematically evaluate LLMs' perspective-taking to identify the most suitable models for a given decision-making task. We discuss the implications for model evaluation. |
| title | A Matter of Perspective(s): Contrasting Human and LLM Argumentation in Subjective Decision-Making on Subtle Sexism |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2502.14052 |