{"type":"ScholarlyArticle","title":"AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents","authors":[{"name":"Petr Anokhin","slug":"petr-anokhin","given":"Petr","family":"Anokhin","affiliation":""},{"name":"Nikita Semenov","slug":"nikita-semenov","given":"Nikita","family":"Semenov","affiliation":""},{"name":"Artyom Sorokin","slug":"artyom-sorokin","given":"Artyom","family":"Sorokin","affiliation":""},{"name":"Dmitry Evseev","slug":"dmitry-evseev","given":"Dmitry","family":"Evseev","affiliation":""},{"name":"Andrey Kravchenko","slug":"andrey-kravchenko","given":"Andrey","family":"Kravchenko","affiliation":""},{"name":"Mikhail Burtsev","slug":"mikhail-burtsev","given":"Mikhail","family":"Burtsev","affiliation":""},{"name":"Evgeny Burnaev","slug":"evgeny-burnaev","given":"Evgeny","family":"Burnaev","affiliation":""}],"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.","doi":"10.24963/ijcai.2025/2","url":"https://ijcai.org/proceedings/2025/2","pdf":"https://ijcai.org/proceedings/2025/0002.pdf","markdown":"https://ijcai.org/proceedings/2025/2.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/2","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Agent-based and Multi-agent Systems","track":"Main Track","number":2,"pages":{"first":12,"last":20},"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"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}