A Self-Healing Framework for Reliable LLM-Based Autonomous Agents

2026-05-07

AI Summary

Large language model (LLM)-driven autonomous agents often exhibit unreliable behaviors in complex software systems due to hallucinations, execution errors, and inconsistent reasoning. To address this challenge, this work proposes the first integrated monitoring framework that jointly models intrinsic reasoning and extrinsic execution. The framework enables precise anomaly detection through a taxonomy of failure types and mechanisms based on execution patterns and output consistency. Furthermore, it introduces a quantitative reliability assessment model that drives dynamic repair strategies, including adaptive replanning and corrective prompting. Experimental results demonstrate that the proposed approach significantly improves task success rates in multi-agent workflows, effectively suppresses failure propagation, and enhances overall system robustness.

Abstract

Autonomous agents based on Large Language Models (LLMs) are increasingly being utilized in complex software systems. However, reliability remains a significant challenge due to unpredictable failures such as hallucinations, execution errors, and inconsistent reasoning. This paper proposes a reliability-aware self-healing framework for LLM-based software agents. The framework integrates failure detection, reliability assessment, and automated recovery mechanisms. First, we define a taxonomy of failure types and introduce a quantitative reliability assessment model. Next, we propose a failure detection method that identifies abnormal agent behavior based on execution patterns and output consistency. Finally, we design a self-healing mechanism that dynamically recovers from failures through adaptive replanning and corrective prompting strategies. The proposed framework was implemented in a multi-agent workflow environment and evaluated using real-world task scenarios. Experimental results demonstrate that our approach significantly increases task success rates, reduces failure propagation, and enhances overall system robustness compared to existing methods. In particular, this study distinguishes itself by establishing an integrated monitoring system that combines the agent's internal reasoning process with external execution results. These findings are expected to contribute to securing the stability of advanced autonomous systems and lowering the barriers to LLM adoption in production environments.

Problem

• reliability
• LLM-based autonomous agents
• failure detection
• self-healing
• hallucinations

Innovation

• self-healing
• reliability assessment
• failure detection
• corrective prompting
• LLM-based agents