AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents

Anokhin, Petr; Semenov, Nikita; Sorokin, Artyom; Evseev, Dmitry; Kravchenko, Andrey; Burtsev, Mikhail; Burnaev, Evgeny

doi:10.24963/ijcai.2025/2

AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents

Petr Anokhin, Nikita Semenov, Artyom Sorokin, Dmitry Evseev, Andrey Kravchenko, Mikhail Burtsev, Evgeny Burnaev

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 12-20.

https://doi.org/10.24963/ijcai.2025/2

Keywords

Natural Language Processing: NLP: Information retrieval and text mining
Agent-based and Multi-agent Systems: MAS: Multi-agent learning
Agent-based and Multi-agent Systems: MAS: Multi-agent planning
Data Mining: DM: Knowledge graphs and knowledge base completion

Abstract

Advancements in the capabilities of Large Language Models (LLMs) have created a promising foundation for developing autonomous agents. With the right tools, these agents could learn to solve tasks in new environments by accumulating and updating their knowledge. Current LLM-based agents process past experiences using a full history of observations, summarization, retrieval augmentation. However, these unstructured memory representations do not facilitate the reasoning and planning essential for complex decision-making. In our study, we introduce AriGraph, a novel method wherein the agent constructs and updates a memory graph that integrates semantic and episodic memories while exploring the environment. We demonstrate that our Ariadne LLM agent, consisting of the proposed memory architecture augmented with planning and decision-making, effectively handles complex tasks within interactive text game environments difficult even for human players. Results show that our approach markedly outperforms other established memory methods and strong RL baselines in a range of problems of varying complexity. Additionally, AriGraph demonstrates competitive performance compared to dedicated knowledge graph-based methods in static multi-hop question-answering.

Petr Anokhin, Nikita Semenov, Artyom Sorokin, Dmitry Evseev, Andrey Kravchenko, Mikhail Burtsev, and Evgeny Burnaev. 2025. AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 12–20. https://doi.org/10.24963/ijcai.2025/2

@inproceedings{ijcai2025p2,
  title     = {AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents},
  author    = {Anokhin, Petr and Semenov, Nikita and Sorokin, Artyom and Evseev, Dmitry and Kravchenko, Andrey and Burtsev, Mikhail and Burnaev, Evgeny},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {12--20},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/2},
  url       = {https://doi.org/10.24963/ijcai.2025/2},
}

TY  - CPAPER
TI  - AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents
AU  - Anokhin, Petr
AU  - Semenov, Nikita
AU  - Sorokin, Artyom
AU  - Evseev, Dmitry
AU  - Kravchenko, Andrey
AU  - Burtsev, Mikhail
AU  - Burnaev, Evgeny
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 12
EP  - 20
DO  - 10.24963/ijcai.2025/2
UR  - https://doi.org/10.24963/ijcai.2025/2
ER  -