Oriel — AI & technology worth noticing, every day.

EvoMemState: Evolving Memory State for Personal Agents

Paper · https://arxiv.org/abs/2611.03743

Long-term memory enables large language model (LLM) agents to provide continuous personalized assistance by retaining user information across interactions. Existing systems support this process through fact extraction, organization, updating, and retrieval, but largely maintain memories as independent facts. This fact-centric paradigm may overlook how new facts update, constrain, or conflict with earlier ones, while leaving their applicability, role, and priority for the current request implicit. In this paper, we introduce EvoMemState, a state-centric framework that augments persistent facts with two complementary states: Relational States that capture how facts remain valid and relate to one another as memory evolves, and a Reasoning State that determines how these facts should be applied to the current request. Specifically, EvoMemState consists of three key components: (i) Ontology-Guided Memory State Induction extracts Persistent Facts and constructs Relational States that maintain their validity, scope, and evolving relations. (ii) RL-Based State Evolution internalizes memory-use experience into a policy through reinforcement learning to generate a request-specific Reasoning State. (iii) State-Aware Reasoning then combines the relevant facts with their Relational States and the generated Reasoning State to guide the personal agent. Extensive experiments on long-term dialogue and memory-use benchmarks demonstrate consistent improvements over existing memory systems in retrieval completeness and memory-grounded response accuracy. Our code and implementation details are available at https://github.com/XMUDeepLIT/EvoMemState.git.

First seen by Oriel on .

AI summary

EvoMemState is a state-centric long-term memory framework that improves LLM personal agents’ retrieval completeness and memory-grounded response accuracy over fact-centric memory systems.

  • Relational States track persistent facts’ validity, scope, constraints, conflicts, and evolving relationships.
  • Ontology-Guided Memory State Induction extracts Persistent Facts and constructs their associated Relational States.
  • RL-Based State Evolution learns a policy from memory-use experience to generate a request-specific Reasoning State.
  • State-Aware Reasoning combines relevant facts, Relational States, and the Reasoning State to guide personalized responses.
  • Experiments report consistent improvements on long-term dialogue and memory-use benchmarks, though the abstract provides no numerical results.

#llm agents #long-term memory #reinforcement learning #personalization

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Points

Score reported by the source.

  • Hugging Face Papers: 31, latest observation ; +11 over observed period since

Comments

Comment count reported by the source.

  • Hugging Face Papers: 5, latest observation ; +2 over observed period since

Aliases

SourceExternal IDURL
Hugging Face Papers2611.03743https://arxiv.org/abs/2611.03743 Read discussion