# Towards In-Depth Root Cause Localization for Microservices with Multi-Agent Recursion-of-Thought

arXiv: https://arxiv.org/abs/2605.14866 · HTML: https://mlsanity.com/paper/2605.14866/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 29 of its 111 references that are in the map.
- First posted on arXiv 14 May 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 29 of 29)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Exploring LLM-based Agents for Root Cause Analysis](https://mlsanity.com/paper/2403.04123/) (2024)
- [OWL: A Large Language Model for IT Operations](https://mlsanity.com/paper/2309.09298/) (2023)
- [A Survey on Parallel Text Generation: From Parallel Decoding to Diffusion Language Models](https://mlsanity.com/paper/2508.08712/) (2025)
- [AgentFM: Role-Aware Failure Management for Distributed Databases with LLM-Driven Multi-Agents](https://mlsanity.com/paper/2504.06614/) (2025)
- [ThinkFL: Self-Refining Failure Localization for Microservice Systems via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2504.18776/) (2025)
- [Omni-SafetyBench: A Benchmark for Safety Evaluation of Audio-Visual Large Language Models](https://mlsanity.com/paper/2508.07173/) (2025)
- [Face It Yourselves: An LLM-Based Two-Stage Strategy to Localize Configuration Errors via Logs](https://mlsanity.com/paper/2404.00640/) (2024)
- [d-TreeRPO: Towards More Reliable Policy Optimization for Diffusion Language Models](https://mlsanity.com/paper/2512.09675/) (2025)
- [LogLM: From Task-based to Instruction-based Automated Log Analysis](https://mlsanity.com/paper/2410.09352/) (2024)
- [TAMO: Fine-Grained Root Cause Analysis via Tool-Assisted LLM Agent with Multi-Modality Observation Data in Cloud-Native Systems](https://mlsanity.com/paper/2504.20462/) (2025)
- [Adaptive Root Cause Localization for Microservice Systems with Multi-Agent Recursion-of-Thought](https://mlsanity.com/paper/2508.20370/) (2025)
- [Cloud Atlas: Efficient Fault Localization for Cloud Systems using Language Models and Causal Insight](https://mlsanity.com/paper/2407.08694/) (2024)
- [MicroRemed: Benchmarking LLMs in Microservices Remediation](https://mlsanity.com/paper/2511.01166/) (2025)
- [Walk the Talk: Is Your Log-based Software Reliability Maintenance System Really Reliable?](https://mlsanity.com/paper/2509.24352/) (2025)
- [LEMMA-RCA: A Large Multi-modal Multi-domain Dataset for Root Cause Analysis](https://mlsanity.com/paper/2406.05375/) (2024)
- [Agentic Memory Enhanced Recursive Reasoning for Root Cause Localization in Microservices](https://mlsanity.com/paper/2601.02732/) (2026)
- [United We Stand: Towards End-to-End Log-based Fault Diagnosis via Interactive Multi-Task Learning](https://mlsanity.com/paper/2509.24364/) (2025)
- [Hypothesize-Then-Verify: Speculative Root Cause Analysis for Microservices with Pathwise Parallelism](https://mlsanity.com/paper/2601.02736/) (2026)
- [LogDB: Multivariate Log-based Failure Diagnosis for Distributed Databases (Extended from MultiLog)](https://mlsanity.com/paper/2505.01676/) (2025)
- [The Multi-Agent Fault Localization System Based on Monte Carlo Tree Search Approach](https://mlsanity.com/paper/2507.22800/) (2025)
- [Efficient Failure Management for Multi-Agent Systems with Reasoning Trace Representation](https://mlsanity.com/paper/2603.21522/) (2026)
- [AnomalyLLM: Few-shot Anomaly Edge Detection for Dynamic Graphs using Large Language Models](https://mlsanity.com/paper/2405.07626/) (2024)
- [GALA: Can Graph-Augmented Large Language Model Agentic Workflows Elevate Root Cause Analysis?](https://mlsanity.com/paper/2508.12472/) (2025)
- [RuntimeSlicer: Towards Generalizable Unified Runtime State Representation for Failure Management](https://mlsanity.com/paper/2603.21495/) (2026)
- [Simplifying Root Cause Analysis in Kubernetes with StateGraph and LLM](https://mlsanity.com/paper/2506.02490/) (2025)
- [E2E-REME: Towards End-to-End Microservices Auto-Remediation via Experience-Simulation Reinforcement Fine-Tuning](https://mlsanity.com/paper/2604.11094/) (2026)
- [Towards Robust LLM Post-Training: Automatic Failure Management for Reinforcement Fine-Tuning](https://mlsanity.com/paper/2605.04431/) (2026)
- [Enhancing Cluster Resilience: LLM-agent Based Autonomous Intelligent Cluster Diagnosis System and Evaluation Framework](https://mlsanity.com/paper/2411.05349/) (2024)

## Built on by (most cited first, 1 of 1)

- [Playing social deduction games with reinforcement fine-tuned large language models](https://mlsanity.com/paper/2610.04261/) (2026)

## Abstract (arXiv)

As modern microservice systems grow increasingly complex due to dynamic interactions and evolving runtime environments, they experience failures with rising frequency. Ensuring system reliability therefore critically depends on accurate root cause localization (RCL). While numerous traditional machine learning and deep learning approaches have been explored for this task, they often suffer from limited interpretability and poor transferability across deployments. More recently, large language model (LLM)-based methods have been proposed to address these issues. However, existing LLM-based approaches still face two fundamental limitations: context explosion, which dilutes critical evidence and degrades localization accuracy, and serial reasoning structures, which hinder deep causal exploration and impair inference efficiency. In this paper, we conduct a comprehensive study of both how human SREs perform root cause localization in practice and why existing LLM-based methods fall short. Motivated by these findings, we introduce RCLAgent, an in-depth root cause localization framework for microservice systems that realizes multi-agent recursion-of-thought with parallel reasoning. RCLAgent decomposes the diagnostic process along the trace graph by assigning each span to a Dedicated Agent and organizing agents recursively and in parallel according to the graph topology, with the final diagnosis obtained by synthesizing the Root-Level Diagnosis Report and the Global Evidence Graph. Extensive experiments on multiple public benchmarks demonstrate that RCLAgent consistently outperforms state-of-the-art methods in both localization accuracy and inference efficiency.