Preprint
GuideLLM: Exploring LLM-Guided Conversation with Applications in Autobiography Interviewing
arXiv.org
10 Feb 2025
Abstract
Although Large Language Models (LLMs) succeed in human-guided conversations such as instruction following and question answering, the potential of LLM-guided conversations-where LLMs direct the discourse and steer the conversation's objectives-remains under-explored. In this study, we first characterize LLM-guided conversation into three fundamental components: (i) Goal Navigation; (ii) Context Management; (iii) Empathetic Engagement, and propose GuideLLM as an installation. We then implement an interviewing environment for the evaluation of LLM-guided conversation. Specifically, various topics are involved in this environment for comprehensive interviewing evaluation, resulting in around 1.4k turns of utterances, 184k tokens, and over 200 events mentioned during the interviewing for each chatbot evaluation. We compare GuideLLM with 6 state-of-the-art LLMs such as GPT-4o and Llama-3-70b-Instruct, from the perspective of interviewing quality, and autobiography generation quality. For automatic evaluation, we derive user proxies from multiple autobiographies and employ LLM-as-a-judge to score LLM behaviors. We further conduct a human-involved experiment by employing 45 human participants to chat with GuideLLM and baselines. We then collect human feedback, preferences, and ratings regarding the qualities of conversation and autobiography. Experimental results indicate that GuideLLM significantly outperforms baseline LLMs in automatic evaluation and achieves consistent leading performances in human ratings.
Metrics
37 Record Views
Details
- Title
- GuideLLM: Exploring LLM-Guided Conversation with Applications in Autobiography Interviewing
- Creators
- Jinhao Duan - Drexel UniversityXinyu Zhao - University of North Carolina at Chapel HillZhuoxuan Zhang - Brown UniversityEunhye Ko - The University of Texas at AustinLily Boddy - The University of Texas at AustinChenan Wang - Drexel UniversityTianhao Li - The University of Texas at AustinAlexander Rasgon - The University of Texas at AustinJunyuan Hong - The University of Texas at AustinMin Kyung Lee - The University of Texas at AustinChenxi Yuan - New Jersey Institute of TechnologyQi Long - University of PennsylvaniaYing Ding - The University of Texas at AustinTianlong Chen - University of North Carolina at Chapel HillKaidi Xu - Drexel University
- Publication Details
- arXiv.org
- Number of pages
- 31
- Resource Type
- Preprint
- Language
- English
- Academic Unit
- Computer Science; Communication
- Other Identifier
- 991022027537104721