From LLM to Conversational Agent: A Memory Enhanced Architecture with Fine-Tuning of Large Language Models

TL;DR

This paper presents RAISE, a memory-augmented architecture that improves the integration of large language models into conversational agents, enhancing context management and adaptability in multi-turn dialogues.

Contribution

It introduces RAISE, a novel framework with a dual-memory system that extends ReAct, enabling more controllable and context-aware conversational agents.

Findings

RAISE improves agent controllability in complex dialogues.

Preliminary results show advantages over traditional agents.

Potential for broader applications in conversational AI.

Abstract

This paper introduces RAISE (Reasoning and Acting through Scratchpad and Examples), an advanced architecture enhancing the integration of Large Language Models (LLMs) like GPT-4 into conversational agents. RAISE, an enhancement of the ReAct framework, incorporates a dual-component memory system, mirroring human short-term and long-term memory, to maintain context and continuity in conversations. It entails a comprehensive agent construction scenario, including phases like Conversation Selection, Scene Extraction, CoT Completion, and Scene Augmentation, leading to the LLMs Training phase. This approach appears to enhance agent controllability and adaptability in complex, multi-turn dialogues. Our preliminary evaluations in a real estate sales context suggest that RAISE has some advantages over traditional agents, indicating its potential for broader applications. This work contributes to the AI field by providing a robust framework for developing more context-aware and versatile conversational agents.