# SAKE: Structured Agentic Knowledge Extrapolation for Complex LLM Reasoning via Reinforcement Learning

arXiv: https://arxiv.org/abs/2505.15062 · HTML: https://mlsanity.com/paper/2505.15062/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 8 of its 22 references that are in the map.
- First posted on arXiv 21 May 2025.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)
- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 8 of 8)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://mlsanity.com/paper/2503.09516/) (2025)
- [ConceptNet 5.5: An Open Multilingual Graph of General Knowledge](https://mlsanity.com/paper/1612.03975/) (2016)
- [Search-o1: Agentic Search-Enhanced Large Reasoning Models](https://mlsanity.com/paper/2501.05366/) (2025)
- [Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph](https://mlsanity.com/paper/2307.07697/) (2023)
- [GIVE: Structured Reasoning of Large Language Models with Knowledge Graph Inspired Veracity Extrapolation](https://mlsanity.com/paper/2410.08475/) (2024)

## Built on by (most cited first, 9 of 9)

- [Enhancing Long Chain-of-Thought Reasoning through Multi-Path Plan Aggregation](https://mlsanity.com/paper/2510.11620/) (2025)
- [Adaptive Information Control for Search-Augmented LLM Reasoning](https://mlsanity.com/paper/2602.01672/) (2026)
- [Stabilizing Off-Policy Training for Long-Horizon LLM Agent via Turn-Level Importance Sampling and Clipping-Triggered Normalization](https://mlsanity.com/paper/2511.20718/) (2025)
- [ETS: Energy-Guided Test-Time Scaling for Training-Free RL Alignment](https://mlsanity.com/paper/2601.21484/) (2026)
- [Anomaly Detection and Early Warning Mechanism for Intelligent Monitoring Systems in Multi-Cloud Environments Based on LLM](https://mlsanity.com/paper/2506.07407/) (2025)
- [Federated Learning-Based Data Collaboration Method for Enhancing Edge Cloud AI System Security Using Large Language Models](https://mlsanity.com/paper/2506.18087/) (2025)
- [E-bike agents: Large Language Model-Driven E-Bike Accident Analysis and Severity Prediction](https://mlsanity.com/paper/2506.04654/) (2025)
- [An Intelligent Fault Self-Healing Mechanism for Cloud AI Systems via Integration of Large Language Models and Deep Reinforcement Learning](https://mlsanity.com/paper/2506.07411/) (2025)
- [Leveraging Large Language Model for Intelligent Log Processing and Autonomous Debugging in Cloud AI Platforms](https://mlsanity.com/paper/2506.17900/) (2025)

## Abstract (arXiv)

Knowledge extrapolation is the process of inferring novel information by combining and extending existing knowledge that is explicitly available. It is essential for solving complex questions in specialized domains where retrieving comprehensive external knowledge is impractical. We propose SAKE (Structured Agentic Knowledge Extrapolation), a RL powered agentic framework that trains LLMs to autonomously retrieve and extrapolate structured knowledge through tool-augmented reinforcement learning. SAKE defines two external KG tools: entity group construction and cross-group triplet retrieval. The model learns to interleave these 2 retrieval tools during a three-turn rollout: extracting key entities, filtering relevant concept groups, and associative reasoning by constructing new triplets through analogy. The entire pipeline is optimized end-to-end with GRPO using a reward that combines output format and answer correctness, teaching the model what to retrieve and how to reason over it. Our experiments proved that SAKE fine-tuned Qwen2.5-7B model surpasses GPT-3.5-Turbo with state-of-the-art agentic KG reasoning on both biomedical (75.4% vs. 70.1%) and commonsense (81.3% vs. 74.7%) benchmarks, while reducing token usage by over 90%. These results demonstrate that associative reasoning over incomplete structured knowledge does not require large models with complex, multi-step prompting, thus can be learned end-to-end by small, open-weight models through reinforcement learning with the right tools and training signal. Our code is available at https://github.com/jxfan99/SAKE.