Reliable Conversational Agents under ASP Control that Understand Natural Language

• DOI: 10.4204/eptcs.416.41
• Publication Date: 2025-02-11T12:05:41Z
• Authors: Yankai Zeng
• Keywords: FOS: Computer and information sciences; Computer Science - Logic in Computer Science; Computer Science - Computation and Language; Computation and Language (cs.CL); Logic in Computer Science (cs.LO)

Efforts have been made to make machines converse like humans in the past few decades. The recent techniques of Large Language Models (LLMs) it possible human-like conversations with machines, but LLM's flaws lacking understanding and reliability are well documented. We believe that best way eliminate this problem is use LLMs only as parsers translate text knowledge vice versa carry out conversation by reasoning over using answer set programming. I developing a framework based on ASP realize reliable chatbots "understand" human conversation. This has used develop task-specific socialbots. My future research focused making these scalable trainable.