Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent
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_ | 1866916524090982400 |
|---|---|
| author | Yu, Xiaoyan Luo, Tongxu Wei, Yifan Lei, Fangyu Huang, Yiming Peng, Hao Zhu, Liehuang |
| author_facet | Yu, Xiaoyan Luo, Tongxu Wei, Yifan Lei, Fangyu Huang, Yiming Peng, Hao Zhu, Liehuang |
| contents | Large Language Models (LLMs) have revolutionized open-domain dialogue agents but encounter challenges in multi-character role-playing (MCRP) scenarios. To address the issue, we present Neeko, an innovative framework designed for efficient multiple characters imitation. Unlike existing methods, Neeko employs a dynamic low-rank adapter (LoRA) strategy, enabling it to adapt seamlessly to diverse characters. Our framework breaks down the role-playing process into agent pre-training, multiple characters playing, and character incremental learning, effectively handling both seen and unseen roles. This dynamic approach, coupled with distinct LoRA blocks for each character, enhances Neeko's adaptability to unique attributes, personalities, and speaking patterns. As a result, Neeko demonstrates superior performance in MCRP over most existing methods, offering more engaging and versatile user interaction experiences. Code and data are available at https://github.com/weiyifan1023/Neeko. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_13717 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent Yu, Xiaoyan Luo, Tongxu Wei, Yifan Lei, Fangyu Huang, Yiming Peng, Hao Zhu, Liehuang Computation and Language Large Language Models (LLMs) have revolutionized open-domain dialogue agents but encounter challenges in multi-character role-playing (MCRP) scenarios. To address the issue, we present Neeko, an innovative framework designed for efficient multiple characters imitation. Unlike existing methods, Neeko employs a dynamic low-rank adapter (LoRA) strategy, enabling it to adapt seamlessly to diverse characters. Our framework breaks down the role-playing process into agent pre-training, multiple characters playing, and character incremental learning, effectively handling both seen and unseen roles. This dynamic approach, coupled with distinct LoRA blocks for each character, enhances Neeko's adaptability to unique attributes, personalities, and speaking patterns. As a result, Neeko demonstrates superior performance in MCRP over most existing methods, offering more engaging and versatile user interaction experiences. Code and data are available at https://github.com/weiyifan1023/Neeko. |
| title | Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2402.13717 |