Larimar: Large Language Models with Episodic Memory Control

TL;DR

Larimar introduces a brain-inspired episodic memory architecture for LLMs that enables fast, accurate, and flexible knowledge updates without retraining, significantly improving efficiency and adaptability.

Contribution

Larimar presents a novel, simple, and LLM-agnostic episodic memory system that allows dynamic knowledge updates, selective forgetting, and context length generalization in large language models.

Findings

Achieves comparable accuracy to baselines in fact editing tasks.

Provides 8-10x speed-up in knowledge updating.

Demonstrates effective mechanisms for forgetting and context management.

Abstract

Efficient and accurate updating of knowledge stored in Large Language Models (LLMs) is one of the most pressing research challenges today. This paper presents Larimar - a novel, brain-inspired architecture for enhancing LLMs with a distributed episodic memory. Larimar's memory allows for dynamic, one-shot updates of knowledge without the need for computationally expensive re-training or fine-tuning. Experimental results on multiple fact editing benchmarks demonstrate that Larimar attains accuracy comparable to most competitive baselines, even in the challenging sequential editing setup, but also excels in speed - yielding speed-ups of 8-10x depending on the base LLM - as well as flexibility due to the proposed architecture being simple, LLM-agnostic, and hence general. We further provide mechanisms for selective fact forgetting, information leakage prevention, and input context length generalization with Larimar and show their effectiveness. Our code is available at https://github.com/IBM/larimar