Reliable Conversational Agents under ASP Control that Understand Natural Language

TL;DR

This paper proposes a framework combining Large Language Models and Answer Set Programming to create reliable conversational agents capable of understanding and reasoning over human language, improving trustworthiness and task-specific performance.

Contribution

The paper introduces a novel framework that integrates LLMs with ASP to enhance reliability and understanding in conversational agents, addressing LLM limitations.

Findings

Framework successfully developed for task-specific chatbots and socialbots.

Improved reliability and understanding in conversations compared to traditional LLM-based chatbots.

Future work aims to enhance scalability and trainability of the system.

Abstract

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