面向社会感知协商对话的辅助大型语言模型智能体
Assistive Large Language Model Agents for Socially-Aware Negotiation Dialogues
AI总结：
该研究开发基于LLMs的辅助智能体助力商务协商，通过角色扮演模拟协商、补救智能体重写违规话语，提出基于价值影响标准的无调优无标签ICL方法识别高质量示例，在三类主题协商中验证有效性并开源相关资源。
摘要：
我们开发了基于大型语言模型（LLMs）的辅助智能体，用于协助对话者进行商务协商。具体而言，我们通过让两个基于LLM的智能体进行角色扮演来模拟商务协商，第三个LLM作为补救智能体，重写违反规范的话语以改善协商结果。我们引入了一种简单的无调优、无标签的上下文学习（ICL）方法，为补救智能体重识别高质量ICL示例，并提出了一种名为价值影响的新型选择标准来衡量协商结果的质量。我们提供了丰富的实证证据，证明其在三个不同协商主题中的有效性，并已在https://github.com/tk1363704/SADAS发布源代码和生成的数据集。
英文摘要：
We develop assistive agents based on Large Language Models (LLMs) that aid interlocutors in business negotiations. Specifically, we simulate business negotiations by letting two LLM-based agents engage in role play. A third LLM acts as a remediator agent to rewrite utterances violating norms for improving negotiation outcomes. We introduce a simple tuning-free and label-free In-Context Learning (ICL) method to identify high-quality ICL exemplars for the remediator, where we propose a novel select criteria, called value impact, to measure the quality of the negotiation outcomes. We provide rich empirical evidence to demonstrate its effectiveness in negotiations across three different negotiation topics. We have released our source code and the generated dataset at: https://github.com/tk1363704/SADAS.
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。
