Personalized LLM Response Generation with Parameterized Memory Injection

2024-04-04

arXiv.org

AI Summary

Large language models (LLMs) lack fine-grained user characteristic modeling for personalized response generation in critical domains such as healthcare. Method: We propose Memory-injected Language Personalization (MiLP), a parametric memory injection framework that overcomes the limited user-detail awareness of conventional memory-augmented approaches. MiLP integrates learnable, parameterized memory injection with Bayesian optimization–driven hyperparameter adaptation, enabling lightweight, dynamic, and customizable personalization. Built upon parameter-efficient fine-tuning (PEFT), it requires no additional inference-time computation. Contribution/Results: MiLP achieves significant improvements in response relevance and consistency across multiple personalized dialogue benchmarks. It supports online user memory updates and introduces zero inference overhead while maintaining high adaptability and efficiency.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large Language Models (LLMs) have exhibited remarkable proficiency in comprehending and generating natural language. On the other hand, personalized LLM response generation holds the potential to offer substantial benefits for individuals in critical areas such as medical. Existing research has explored memory-augmented methods to prompt the LLM with pre-stored user-specific knowledge for personalized response generation in terms of new queries. We contend that such paradigm is unable to perceive fine-granularity information. In this study, we propose a novel extbf{M}emory- extbf{i}njected approach using parameter-efficient fine-tuning (PEFT) and along with a Bayesian Optimisation searching strategy to achieve extbf{L}LM extbf{P}ersonalization( extbf{MiLP}).

Problem

• Memory Enhancement
• Personalization
• Large Language Models

Innovation

• MiLP
• Bayesian Optimization
• Personalized Responses