IntentGuide: Neuro-Symbolic Integration for Clarifying Human Intents by Rewriting Free-Form Sentences

This paper presents “IntentGuide,” a neuro-symbolic integration framework to enhance the clarity and executability of human intentions expressed in free-form sentences. IntentGuide effectively integrates the rule-based error detection capabilities of symbolic AI with the powerful adaptive learning abilities of Large Language Model (LLM) to convert ambiguous and/or complex sentences into clear, machine-understandable English instructions. The empirical evaluation of IntentGuide, performed on natural language sentences written by middle school students for designing video games, reveals a significant improvement in error correction and code generation abilities compared to previous approaches, attaining an accuracy rate of 90%.