LLM-based Medical Assistant Personalization with Short- and Long-Term Memory Coordination

TL;DR

This paper introduces a resource-efficient method for personalizing medical assistants using a novel memory mechanism and parameter-efficient fine-tuning of large language models, enhancing user-specific responses without full model retraining.

Contribution

It proposes a new bionic memory mechanism combined with PEFT to personalize LLM-based medical assistants efficiently.

Findings

Demonstrates improved personalization in medical assistants.

Reduces resource consumption compared to full model training.

Enhances response accuracy with memory coordination.

Abstract

Large Language Models (LLMs), such as GPT3.5, have exhibited remarkable proficiency in comprehending and generating natural language. On the other hand, medical assistants hold the potential to offer substantial benefits for individuals. However, the exploration of LLM-based personalized medical assistant remains relatively scarce. Typically, patients converse differently based on their background and preferences which necessitates the task of enhancing user-oriented medical assistant. While one can fully train an LLM for this objective, the resource consumption is unaffordable. Prior research has explored memory-based methods to enhance the response with aware of previous mistakes for new queries during a dialogue session. We contend that a mere memory module is inadequate and fully training an LLM can be excessively costly. In this study, we propose a novel computational bionic memory mechanism, equipped with a parameter-efficient fine-tuning (PEFT) schema, to personalize medical assistants.