KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents

2024-03-05

arXiv.org

AI Summary

To address planning hallucinations in large language model (LLM) agents arising from insufficient explicit action knowledge during environmental interaction, this paper proposes a knowledge-enhanced planning framework. The method introduces (1) the first action knowledge base tailored for interactive tasks, explicitly encoding executable actions and their preconditions/constraints; and (2) a knowledge-driven self-iterative planning mechanism that dynamically refines planning trajectories via multi-stage trajectory synthesis and prompt-coordinated optimization. Evaluated on HotpotQA and ALFWorld, the approach matches or surpasses state-of-the-art baselines, achieving up to 12.3% improvement in planning accuracy and an average 37.6% reduction in hallucination rate. It effectively mitigates semantic-action mismatch—a core challenge in embodied reasoning—by grounding LLM-generated plans in executable, constraint-aware action semantics.

Technology Category

• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Planning with Language Models

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: 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

Large Language Models (LLMs) have demonstrated great potential in complex reasoning tasks, yet they fall short when tackling more sophisticated challenges, especially when interacting with environments through generating executable actions. This inadequacy primarily stems from the lack of built-in action knowledge in language agents, which fails to effectively guide the planning trajectories during task solving and results in planning hallucination. To address this issue, we introduce KnowAgent, a novel approach designed to enhance the planning capabilities of LLMs by incorporating explicit action knowledge. Specifically, KnowAgent employs an action knowledge base and a knowledgeable self-learning strategy to constrain the action path during planning, enabling more reasonable trajectory synthesis, and thereby enhancing the planning performance of language agents. Experimental results on HotpotQA and ALFWorld based on various backbone models demonstrate that KnowAgent can achieve comparable or superior performance to existing baselines. Further analysis indicates the effectiveness of KnowAgent in terms of planning hallucinations mitigation. Code is available in https://github.com/zjunlp/KnowAgent.

Problem

• Large Language Models
• Practical Action Knowledge
• Task Execution Errors

Innovation

• KnowAgent
• Action Knowledge Base
• Self-Learning Mechanism