ERABAL: Enhancing Role-Playing Agents through Boundary-Aware Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910659599400960 |
|---|---|
| author | Tang, Yihong Ou, Jiao Liu, Che Zhang, Fuzheng Zhang, Di Gai, Kun |
| author_facet | Tang, Yihong Ou, Jiao Liu, Che Zhang, Fuzheng Zhang, Di Gai, Kun |
| contents | Role-playing is an emerging application in the field of Human-Computer Interaction (HCI), primarily implemented through the alignment training of a large language model (LLM) with assigned characters. Despite significant progress, role-playing agents (RPLAs) still struggle with maintaining role-consistency across conversations, particularly when confronted with boundary queries subtly related to character attributes. In this paper, we present ERABAL, a framework aimed at enhancing RPLAs' role-playing capabilities through boundary-aware learning. ERABAL encompasses a generation pipeline for role-specific dialogues and a concomitant methodology for alignment training. Through comprehensive evaluations, we demonstrate that ERABAL is both efficient and effective. By training with significantly fewer dialogues than those used in leading approaches, ERABAL achieves notable improvements across WikiRoleEval, CharacterEval, and the role-playing subset of MT-Bench compared to the generalist baseline models. Our code and datasets will be made publicly available to support further research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_14710 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ERABAL: Enhancing Role-Playing Agents through Boundary-Aware Learning Tang, Yihong Ou, Jiao Liu, Che Zhang, Fuzheng Zhang, Di Gai, Kun Computation and Language Artificial Intelligence Role-playing is an emerging application in the field of Human-Computer Interaction (HCI), primarily implemented through the alignment training of a large language model (LLM) with assigned characters. Despite significant progress, role-playing agents (RPLAs) still struggle with maintaining role-consistency across conversations, particularly when confronted with boundary queries subtly related to character attributes. In this paper, we present ERABAL, a framework aimed at enhancing RPLAs' role-playing capabilities through boundary-aware learning. ERABAL encompasses a generation pipeline for role-specific dialogues and a concomitant methodology for alignment training. Through comprehensive evaluations, we demonstrate that ERABAL is both efficient and effective. By training with significantly fewer dialogues than those used in leading approaches, ERABAL achieves notable improvements across WikiRoleEval, CharacterEval, and the role-playing subset of MT-Bench compared to the generalist baseline models. Our code and datasets will be made publicly available to support further research. |
| title | ERABAL: Enhancing Role-Playing Agents through Boundary-Aware Learning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2409.14710 |