arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.13752cs.CLcs.AI
收藏

思考后发言:一个带有思维状态的角色扮演模型

Thinking Before Speaking: A Role-playing Model with Mindset

Beijing Institute of Technology(北京理工大学)

Baohua Zhang, Yongyi Huang, Wenyao Cui, Huaping Zhang

AI总结:

针对LLM角色扮演易被识破且难以处理角色知识盲区的问题,提出TBS模型,通过在对话数据中补充角色思维状态并引入超纲数据微调,使模型在语气、知识和思维上更逼真地模拟真实角色。

摘要:

角色扮演对大型语言模型(LLMs)来说是一项简单的任务,因为它们擅长模拟人类行为。当前的许多研究通过微调模型或使用专门的提示词,使LLMs能够以特定角色的语气生成回复。然而,通常很容易识别出角色是由LLM扮演的。当面对所扮演角色不具备的知识,或需要该角色的特定经验或逻辑来回答的问题时,这些模型往往表现不佳。为了解决这个问题并使LLMs表现得更像真实角色,我们在本文中提出了一个“思考后发言”(Thinking Before Speaking, TBS)模型。与其他研究不同,我们首先基于角色的现实生活场景和历史对话扩展数据,为每一对对话补充角色的思维状态。然后,我们添加了少量包含角色知识范围之外元素的数据点,并对LLMs进行微调。这种方法可以帮助LLMs采用角色的思考过程和逻辑,避免产生超出角色知识库的回复。我们还准备了一个数据集和评估指标来测试这些能力。实验结果表明,我们的TBS模型在语气、知识和思维状态方面能更好地模拟角色。

英文摘要:

Role-playing is an easy task for Large Language Models (LLMs), as they are skilled at simulating human behaviors. Many current studies have enabled LLMs to generate responses in the tone of a specific role by fine-tuning the models or using specialized prompts. However, it is typically easy to recognize when a role is being played by LLMs. These models tend to perform poorly when confronted with knowledge that the assumed role does not possess, or a question that requires the specific experience or logic of the role to answer. To address this problem and make LLMs act more like real roles, we propose a Thinking Before Speaking (TBS) model in this paper. Unlike other studies, we first extend the data based on the character's real-life scenarios and the historical dialogue, supplementing each pair of dialogue with the character's mindset. Then we add few data points that include elements beyond the role's knowledge, and fine-tune the LLMs. This approach can help LLMs adopt the role's thought process and logic, avoiding responses that fall outside the role's knowledge base. We have also prepared a dataset and evaluation metrics to test these capabilities. Experimental results show that our TBS model can better emulate a role in terms of tone, knowledge, and mindset.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑