Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

Abstract

In this paper, we propose MPC (Modular Prompted Chatbot), a new approach for creating high-quality conversational agents without the need for fine-tuning. Our method utilizes pre-trained large language models (LLMs) as individual modules for long-term consistency and flexibility, by using techniques such as few-shot prompting, chain-of-thought (CoT), and external memory. Our human evaluation results show that MPC is on par with fine-tuned chatbot models in open-domain conversations, making it an effective solution for creating consistent and engaging chatbots.

Accepted to the Findings of ACL2023. The camera-ready version with additional experimental results will be uploaded

References in corpus (9)

• PaLM: Scaling Language Modeling with Pathways
• Scaling Instruction-Finetuned Language Models
• Large Language Models are Zero-Shot Reasoners
• Emergent Abilities of Large Language Models
• LaMDA: Language Models for Dialog Applications
• Decomposed Prompting: A Modular Approach for Solving Complex Tasks
• Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP
• Beyond Goldfish Memory: Long-Term Open-Domain Conversation
• Teaching Algorithmic Reasoning via In-context Learning