Prompting and evaluating large language models for proactive dialogues: Clarification, target-guided, and non-collaboration

• Main Authors: DENG, Yang, LIAO, Lizi, CHEN, Liang, WANG, Hongru, LEI, Wenqiang, CHUA, Tat-Seng
• 格式: text
• 語言: English
• 出版: Institutional Knowledge at Singapore Management University 2023
• 主題: Comprehensive analysis; Conversational agents; Conversational systems; Empirical findings; In contexts; Language model; Model-based OPC; Planning capability; Proactivity; Response generation; Databases and Information Systems; Information Security
• 在線閱讀: https://ink.library.smu.edu.sg/sis_research/9116; https://ink.library.smu.edu.sg/context/sis_research/article/10119/viewcontent/Prompting.pdf
• 機構: Singapore Management University

Conversational systems based on Large Language Models (LLMs), such as ChatGPT, show exceptional proficiency in context understanding and response generation. However, they still possess limitations, such as failing to ask clarifying questions to ambiguous queries or refuse users' unreasonable requests, both of which are considered as key aspects of a conversational agent's proactivity. This raises the question of whether LLM-based conversational systems are equipped to handle proactive dialogue problems. In this work, we conduct a comprehensive analysis of LLM-based conversational systems, specifically focusing on three key aspects of proactive dialogues: clarification, target-guided, and non-collaborative dialogues. To trigger the proactivity of LLMs, we propose the Proactive Chain-of-Thought prompting scheme, which augments LLMs with the goal planning capability over descriptive reasoning chains. Empirical findings are discussed to promote future studies on LLM-based proactive dialogue systems.