Reverse Thinking Enhances Missing Information Detection in Large Language Models

2025-12-10

AI Summary

Large language models (LLMs) frequently generate incomplete responses, factual errors, and hallucinations under information-deficient conditions; existing forward-reasoning methods (e.g., Chain-of-Thought, Tree-of-Thought) lack systematic mechanisms to detect missing premises. Method: We propose the first backward-reasoning framework explicitly designed for missing-information detection—leveraging goal-directed conditional backtracking, necessity verification, and multi-step attribution to shift the paradigm from “answer generation” to “premise validation.” Implemented via prompt engineering, it requires no fine-tuning or additional training. Contribution/Results: Evaluated across multiple missing-reasoning benchmarks, our method achieves an average 32.7% improvement in missing-premise identification accuracy and reduces hallucination rates by 41.5%, significantly outperforming CoT and ToT. This work pioneers the systematic application of backward reasoning to LLM-based diagnostic tasks for missing information.

Technology Category

Knowledge Representation and Reasoning: Diagnosis and Abductive ReasoningCognitive Modeling & Cognitive Systems: Conceptual Inference and ReasoningReasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved informationSemantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsUser Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLMs) have demonstrated remarkable capabilities in various reasoning tasks, yet they often struggle with problems involving missing information, exhibiting issues such as incomplete responses, factual errors, and hallucinations. While forward reasoning approaches like Chain-of-Thought (CoT) and Tree-of-Thought (ToT) have shown success in structured problem-solving, they frequently fail to systematically identify and recover omitted information. In this paper, we explore the potential of reverse thinking methodologies to enhance LLMs' performance on missing information detection tasks. Drawing inspiration from recent work on backward reasoning, we propose a novel framework that guides LLMs through reverse thinking to identify necessary conditions and pinpoint missing elements. Our approach transforms the challenging task of missing information identification into a more manageable backward reasoning problem, significantly improving model accuracy. Experimental results demonstrate that our reverse thinking approach achieves substantial performance gains compared to traditional forward reasoning methods, providing a promising direction for enhancing LLMs' logical completeness and reasoning robustness.

Problem

• Enhances LLMs' detection of missing information
• Transforms missing information identification into backward reasoning
• Improves logical completeness and reasoning robustness in LLMs

Innovation

• Reverse thinking framework for missing information detection
• Transforms identification into backward reasoning problem
• Guides LLMs to pinpoint missing elements via necessary conditions