Position: Episodic Memory is the Missing Piece for Long-Term LLM Agents

TL;DR

The paper argues that turning to episodic memory is essential for enabling LLM agents to operate effectively over long timescales. It operationalizes episodic memory for LLMs by defining five key properties that separate it from other memory types and proposes a unifying framework that integrates in-context, external, and parametric memory through a Complementary Learning Systems lens. A concrete roadmap (encoding, retrieval, consolidation, benchmarks) and a set of research questions are offered to drive progress toward long-term, context-sensitive agents. The work emphasizes the practical impact of robust episodic memory for continual learning, knowledge consolidation, and scalable long-term reasoning in AI systems.

Abstract

As Large Language Models (LLMs) evolve from text-completion tools into fully fledged agents operating in dynamic environments, they must address the challenge of continually learning and retaining long-term knowledge. Many biological systems solve these challenges with episodic memory, which supports single-shot learning of instance-specific contexts. Inspired by this, we present an episodic memory framework for LLM agents, centered around five key properties of episodic memory that underlie adaptive and context-sensitive behavior. With various research efforts already partially covering these properties, this position paper argues that now is the right time for an explicit, integrated focus on episodic memory to catalyze the development of long-term agents. To this end, we outline a roadmap that unites several research directions under the goal to support all five properties of episodic memory for more efficient long-term LLM agents.

Figures (1)

• Figure 1: LLM-Agents with an Episodic Memory system. The LLM agent acts on and gets feedback from an environment. Feedback can come in the form of outputs from programs (E1), from other agents (E2), humans (E3), as well as external real-world data (E4). Actions can modify parts of the environment, and provide feedback for humans or other agents in the environment. Within the agent, an external memory system acts as a bridge between parametric and in-context memory while allowing for fast encoding of and retrieval into in-context memory (the LLM's context window). (a) Consolidation: Episodes in the external memory are consolidated into a model's broader parametric memory to avoid capacity limitations and allow for generalization to new semantic knowledge and procedural skills based on specific instances. (b) Encoding: Limited in-context memory can offload its content into external memory. (c) Retrieval: Stored episodes can later be retrieved and used to reinstate representations into in-context memory.