IAO Prompting: Making Knowledge Flow Explicit in LLMs through Structured Reasoning Templates

2025-02-05

AI Summary

Current large language models (LLMs) exhibit implicit and non-auditable knowledge invocation during complex reasoning, resulting in poor factual consistency, undetectable hallucinations, and weak interpretability. To address this, we propose the Input-Action-Output (IAO) prompting framework—the first to enable explicit modeling and end-to-end tracing of knowledge flow by decomposing reasoning into three structured phases: input knowledge, executable action, and output result. Our method integrates templated prompting with zero-shot inference, supporting knowledge gap identification, factual consistency verification, and hallucination detection. On multi-task benchmarks, IAO achieves significant gains in zero-shot accuracy. Human evaluation confirms its effectiveness in localizing reasoning errors. Crucially, it delivers both performance improvements and strong interpretability, establishing a verifiable and auditable paradigm for knowledge utilization in LLMs.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

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

Abstract

While Large Language Models (LLMs) demonstrate impressive reasoning capabilities, understanding and validating their knowledge utilization remains challenging. Chain-of-thought (CoT) prompting partially addresses this by revealing intermediate reasoning steps, but the knowledge flow and application remain implicit. We introduce IAO (Input-Action-Output) prompting, a structured template-based method that explicitly models how LLMs access and apply their knowledge during complex reasoning tasks. IAO decomposes problems into sequential steps, each clearly identifying the input knowledge being used, the action being performed, and the resulting output. This structured decomposition enables us to trace knowledge flow, verify factual consistency, and identify potential knowledge gaps or misapplications. Through experiments across diverse reasoning tasks, we demonstrate that IAO not only improves zero-shot performance but also provides transparency in how LLMs leverage their stored knowledge. Human evaluation confirms that this structured approach enhances our ability to verify knowledge utilization and detect potential hallucinations or reasoning errors. Our findings provide insights into both knowledge representation within LLMs and methods for more reliable knowledge application.

Problem

• Explicitly model knowledge flow in LLMs
• Improve transparency in LLM reasoning tasks
• Verify knowledge utilization and detect errors

Innovation

• IAO Prompting
• Structured Reasoning Templates
• Explicit Knowledge Flow