Identifying and Answering Questions with False Assumptions: An Interpretable Approach

2025-08-20

AI Summary

This work addresses the hallucination problem exhibited by large language models (LLMs) when answering questions containing false premises. Methodologically, it decomposes each input question into multiple verifiable atomic hypotheses and employs an external evidence retrieval module coupled with a fact-checking component to independently validate each hypothesis—thereby detecting and localizing erroneous premises. The key contribution is the first formulation of question answering as an atomic hypothesis verification task, enabling end-to-end interpretable verification. Experiments across five state-of-the-art LLMs demonstrate that the framework significantly reduces hallucination rates (average reduction of 32.7%), improves answer accuracy and premise identification precision, and provides transparent, stepwise attribution paths. This enhances both user trust and model debuggability through explicit, traceable reasoning.

Technology Category

• Natural Language Processing: Question Answering
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

People often ask questions with false assumptions, a type of question that does not have regular answers. Answering such questions require first identifying the false assumptions. Large Language Models (LLMs) often generate misleading answers because of hallucinations. In this paper, we focus on identifying and answering questions with false assumptions in several domains. We first investigate to reduce the problem to fact verification. Then, we present an approach leveraging external evidence to mitigate hallucinations. Experiments with five LLMs demonstrate that (1) incorporating retrieved evidence is beneficial and (2) generating and validating atomic assumptions yields more improvements and provides an interpretable answer by specifying the false assumptions.

Problem

• Identifying false assumptions in questions
• Mitigating hallucinations in LLM responses
• Providing interpretable answers with evidence

Innovation

• Leveraging external evidence to mitigate hallucinations
• Reducing problem to fact verification process
• Generating and validating atomic assumptions interpretably