AI RESEARCH
A Hierarchical Error-Corrective Graph Framework for Autonomous Agents with LLM-Based Action Generation
arXiv CS.AI
•
ArXi:2603.08388v1 Announce Type: new We propose a Hierarchical Error-Corrective Graph FrameworkforAutonomousAgentswithLLM-BasedActionGeneration(HECG),whichincorporates three core innovations: (1) Multi-Dimensional Transferable Strategy (MDTS): by integrating task quality metrics (Q), confidence/cost metrics (C), reward metrics (R), and LLM-based semantic reasoning scores (LLM-Score), MDTS achieves multi-dimensional alignment between quantitative performance and semantic context, enabling precise selection of high-quality candidate strate gies and effectively reducing the risk of negative.