Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2410.03181 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916422922272768 |
|---|---|
| author | Sun, Seungjong Lee, Eungu Baek, Seo Yeon Hwang, Seunghyun Lee, Wonbyung Nan, Dongyan Jansen, Bernard J. Kim, Jang Hyun |
| author_facet | Sun, Seungjong Lee, Eungu Baek, Seo Yeon Hwang, Seunghyun Lee, Wonbyung Nan, Dongyan Jansen, Bernard J. Kim, Jang Hyun |
| contents | This study is the first to explore whether multi-modal large language models (LLMs) can align their behaviors with visual personas, addressing a significant gap in the literature that predominantly focuses on text-based personas. We developed a novel dataset of 5K fictional avatar images for assignment as visual personas to LLMs, and analyzed their negotiation behaviors based on the visual traits depicted in these images, with a particular focus on aggressiveness. The results indicate that LLMs assess the aggressiveness of images in a manner similar to humans and output more aggressive negotiation behaviors when prompted with an aggressive visual persona. Interestingly, the LLM exhibited more aggressive negotiation behaviors when the opponent's image appeared less aggressive than their own, and less aggressive behaviors when the opponents image appeared more aggressive. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_03181 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Kiss up, Kick down: Exploring Behavioral Changes in Multi-modal Large Language Models with Assigned Visual Personas Sun, Seungjong Lee, Eungu Baek, Seo Yeon Hwang, Seunghyun Lee, Wonbyung Nan, Dongyan Jansen, Bernard J. Kim, Jang Hyun Computation and Language This study is the first to explore whether multi-modal large language models (LLMs) can align their behaviors with visual personas, addressing a significant gap in the literature that predominantly focuses on text-based personas. We developed a novel dataset of 5K fictional avatar images for assignment as visual personas to LLMs, and analyzed their negotiation behaviors based on the visual traits depicted in these images, with a particular focus on aggressiveness. The results indicate that LLMs assess the aggressiveness of images in a manner similar to humans and output more aggressive negotiation behaviors when prompted with an aggressive visual persona. Interestingly, the LLM exhibited more aggressive negotiation behaviors when the opponent's image appeared less aggressive than their own, and less aggressive behaviors when the opponents image appeared more aggressive. |
| title | Kiss up, Kick down: Exploring Behavioral Changes in Multi-modal Large Language Models with Assigned Visual Personas |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2410.03181 |