# R-Log: Incentivizing Log Analysis Capability in LLMs via Reasoning-based Reinforcement Learning

arXiv: https://arxiv.org/abs/2509.25987 · HTML: https://mlsanity.com/paper/2509.25987/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 19 of its 72 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 19 of 19)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Better Zero-Shot Reasoning with Role-Play Prompting](https://mlsanity.com/paper/2308.07702/) (2023)
- [S*: Test Time Scaling for Code Generation](https://mlsanity.com/paper/2502.14382/) (2025)
- [Fin-R1: A Large Language Model for Financial Reasoning through Reinforcement Learning](https://mlsanity.com/paper/2503.16252/) (2025)
- [Loghub: A Large Collection of System Log Datasets for AI-driven Log Analytics](https://mlsanity.com/paper/2008.06448/) (2020)
- [R1-T1: Fully Incentivizing Translation Capability in LLMs via Reasoning Learning](https://mlsanity.com/paper/2502.19735/) (2025)
- [ThinkFL: Self-Refining Failure Localization for Microservice Systems via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2504.18776/) (2025)
- [UniLog: Deploy One Model and Specialize it for All Log Analysis Tasks](https://mlsanity.com/paper/2112.03159/) (2021)
- [LLMs Faithfully and Iteratively Compute Answers During CoT: A Systematic Analysis With Multi-step Arithmetics](https://mlsanity.com/paper/2412.01113/) (2024)
- [Domaino1s: Guiding LLM Reasoning for Explainable Answers in High-Stakes Domains](https://mlsanity.com/paper/2501.14431/) (2025)
- [Summarizing Unstructured Logs in Online Services](https://mlsanity.com/paper/2012.08938/) (2020)
- [Taming Text-to-Image Synthesis for Novices: User-centric Prompt Generation via Multi-turn Guidance](https://mlsanity.com/paper/2408.12910/) (2024)
- [Enhancing Interpretability in Software Change Management with Chain-of-Thought Reasoning](https://mlsanity.com/paper/2507.09315/) (2025)

## Built on by (most cited first, 3 of 3)

- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [Chart Specification: Structural Representations for Incentivizing VLM Reasoning in Chart-to-Code Generation](https://mlsanity.com/paper/2602.10880/) (2026)
- [OpsMem: Dual-Memory Reasoning with Cross-Memory Resonance for Failure Diagnosis](https://mlsanity.com/paper/2607.11357/) (2026)

## Abstract (arXiv)

The growing complexity of log data in modern software systems has prompted the use of Large Language Models (LLMs) for automated log analysis. Current approaches typically rely on direct supervised fine-tuning (SFT) on log-label pairs. However, this exacerbates the domain discrepancy between general-purpose LLMs and specialized log data, causing overfitting. Furthermore, SFT's imbalanced loss computation often allows lengthy contexts to overwhelm critical, concise details in model answers, leading to hallucinations. To address these limitations, we propose R-Log, a novel reasoning-based paradigm that mirrors the structured, step-by-step analytical process of human engineers. This approach enhances generalizability by learning the underlying rules behind conclusions. We further employ Reinforcement Learning (RL) to optimize the model within a simulated O&M environment, thereby reducing hallucinations by directly rewarding correct outcomes. R-Log is first cold-started on a curated dataset of 2k+ reasoning trajectories, guided by 13 strategies from manual O&M practices, to establish an initial reasoning capability. This ability is then refined via RL using a joint reward function. Empirical evaluations on real-world logs show that R-Log outperforms existing methods across five log analysis tasks, particularly in unseen scenarios (by 228.05%). We also designed R-Log-fast with 5x speedup while keeping 93% of the efficacy.