Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.07698 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912319601115136 |
|---|---|
| author | Sato, Shiki Baba, Jun Hentona, Asahi Iwata, Shinji Yoshimoto, Akifumi Yoshino, Koichiro |
| author_facet | Sato, Shiki Baba, Jun Hentona, Asahi Iwata, Shinji Yoshimoto, Akifumi Yoshino, Koichiro |
| contents | Chat-oriented dialogue systems designed to provide tangible benefits, such as sharing the latest news or preventing frailty in senior citizens, often require Proactive acquisition of specific user Information via chats on user-faVOred Topics (PIVOT). This study proposes the PIVOT task, designed to advance the technical foundation for these systems. In this task, a system needs to acquire the answers of a user to predefined questions without making the user feel abrupt while engaging in a chat on a predefined topic. We found that even recent large language models (LLMs) show a low success rate in the PIVOT task. We constructed a dataset suitable for the analysis to develop more effective systems. Finally, we developed a simple but effective system for this task by incorporating insights obtained through the analysis of this dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_07698 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Proactive User Information Acquisition via Chats on User-Favored Topics Sato, Shiki Baba, Jun Hentona, Asahi Iwata, Shinji Yoshimoto, Akifumi Yoshino, Koichiro Computation and Language Chat-oriented dialogue systems designed to provide tangible benefits, such as sharing the latest news or preventing frailty in senior citizens, often require Proactive acquisition of specific user Information via chats on user-faVOred Topics (PIVOT). This study proposes the PIVOT task, designed to advance the technical foundation for these systems. In this task, a system needs to acquire the answers of a user to predefined questions without making the user feel abrupt while engaging in a chat on a predefined topic. We found that even recent large language models (LLMs) show a low success rate in the PIVOT task. We constructed a dataset suitable for the analysis to develop more effective systems. Finally, we developed a simple but effective system for this task by incorporating insights obtained through the analysis of this dataset. |
| title | Proactive User Information Acquisition via Chats on User-Favored Topics |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.07698 |