Right for Right Reasons: Large Language Models for Verifiable Commonsense Knowledge Graph Question Answering

2024-03-03

Conference on Empirical Methods in Natural Language Processing

AI Summary

Existing LLM-based knowledge graph question answering (KGQA) methods suffer from hallucination—especially on commonsense reasoning involving long-tail entities and counterfactual questions—and produce unverifiable reasoning traces. This paper proposes R³, a verifiable reasoning paradigm that explicitly axiomatizes implicit commonsense knowledge in LLMs and enforces strict grounding of each inference step in KG triplets, unifying correctness with interpretability. R³ integrates prompt engineering, triplet retrieval, and constrained reasoning chain generation within a dual-track KG–LLM collaborative framework. Evaluated on commonsense QA, claim verification, and preference matching, R³ reduces hallucination rate by 42% and reasoning errors by 37%, while enabling end-to-end traceable verification. It establishes the first formal, commonsense-driven, and verifiably sound KGQA paradigm.

Technology Category

• Knowledge Representation and Reasoning: Common-Sense Reasoning
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Knowledge Graph Question Answering (KGQA) methods seek to answer Natural Language questions using the relational information stored in Knowledge Graphs (KGs). With the recent advancements of Large Language Models (LLMs) and their remarkable reasoning abilities, there is a growing trend to leverage them for KGQA. However, existing methodologies have only focused on answering factual questions, e.g., *“In which city was Silvio Berlusconi’s first wife born?”*, leaving questions involving commonsense reasoning that real-world users may pose more often, e.g., *“Do I need separate visas to see the Venus of Willendorf and attend the Olympics this summer?”* unaddressed. In this work, we first observe that existing LLM-based methods for KGQA struggle with hallucination on such questions, especially on queries targeting long-tail entities (e.g., non-mainstream and recent entities), thus hindering their applicability in real-world applications especially since their reasoning processes are not easily verifiable. In response, we propose Right for Right Reasons (R^3), a commonsense KGQA methodology that allows for a verifiable reasoning procedure by axiomatically surfacing intrinsic commonsense knowledge of LLMs and grounding every factual reasoning step on KG triples. Through experimental evaluations across three different tasks—question answering, claim verification, and preference matching—our findings showcase R^3 as a superior approach, outperforming existing methodologies and notably reducing instances of hallucination and reasoning errors.

Problem

• Addresses hallucination in LLM-based KGQA for commonsense questions
• Improves reasoning verifiability by grounding facts on KG triples
• Enhances performance on long-tail entities and real-world queries

Innovation

• Uses LLMs for verifiable commonsense KGQA
• Grounds reasoning steps on KG triples
• Reduces hallucination in long-tail queries