Hello Again! LLM-powered Personalized Agent for Long-term Dialogue

2024-06-09

arXiv.org

AI Summary

Existing open-domain dialogue systems primarily focus on single-turn interactions, lacking cross-session memory and dynamic personality modeling capabilities—thus failing to support long-term companion-style engagement. To address this, we propose LD-Agent, a personalized agent designed for longitudinal dialogue. LD-Agent introduces a model-agnostic, three-module decoupled architecture: event perception, personality extraction, and response generation. It incorporates a dual-timescale memory separation mechanism, topic-guided retrieval, and a user-agent co-adaptive dynamic personality encoder. Leveraging an LLM-driven event memory bank, LD-Agent enables fine-grained event summarization and role evolution modeling. Extensive evaluations across multiple benchmarks, foundation models, and downstream tasks demonstrate LD-Agent’s effectiveness and strong cross-domain generalizability. The implementation is publicly available.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Natural Language Processing: Conversational AI/Dialog Systems

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: Agentic search

Abstract

Open-domain dialogue systems have seen remarkable advancements with the development of large language models (LLMs). Nonetheless, most existing dialogue systems predominantly focus on brief single-session interactions, neglecting the real-world demands for long-term companionship and personalized interactions with chatbots. Crucial to addressing this real-world need are event summary and persona management, which enable reasoning for appropriate long-term dialogue responses. Recent progress in the human-like cognitive and reasoning capabilities of LLMs suggests that LLM-based agents could significantly enhance automated perception, decision-making, and problem-solving. In response to this potential, we introduce a model-agnostic framework, the Long-term Dialogue Agent (LD-Agent), which incorporates three independently tunable modules dedicated to event perception, persona extraction, and response generation. For the event memory module, long and short-term memory banks are employed to separately focus on historical and ongoing sessions, while a topic-based retrieval mechanism is introduced to enhance the accuracy of memory retrieval. Furthermore, the persona module conducts dynamic persona modeling for both users and agents. The integration of retrieved memories and extracted personas is subsequently fed into the generator to induce appropriate responses. The effectiveness, generality, and cross-domain capabilities of LD-Agent are empirically demonstrated across various illustrative benchmarks, models, and tasks. The code is released at https://github.com/leolee99/LD-Agent.

Problem

• Enhances long-term dialogue interactions
• Improves personalized chatbot responses
• Manages event summaries and personas effectively

Innovation

• LLM-based long-term dialogue agent
• Event memory and persona modules
• Topic-based retrieval mechanism