HabitatAgent: An End-to-End Multi-Agent System for Housing Consultation

TL;DR

HabitatAgent is a novel multi-agent system powered by large language models designed to provide transparent, reliable, and end-to-end housing consultation by integrating specialized agents for memory, retrieval, generation, and validation.

Contribution

This work introduces HabitatAgent, the first multi-agent LLM-powered architecture for housing consultation, improving accuracy and transparency over existing recommendation systems.

Findings

HabitatAgent achieves 95% accuracy on real user scenarios.

Baseline Dense+Rerank achieves 75% accuracy.

HabitatAgent provides an auditable, end-to-end decision workflow.

Abstract

Housing selection is a high-stakes and largely irreversible decision problem. We study housing consultation as a decision-support interface for housing selection. Existing housing platforms and many LLM-based assistants often reduce this process to ranking or recommendation, resulting in opaque reasoning, brittle multi-constraint handling, and limited guarantees on factuality. We present HabitatAgent, the first LLM-powered multi-agent architecture for end-to-end housing consultation. HabitatAgent comprises four specialized agent roles: Memory, Retrieval, Generation, and Validation. The Memory Agent maintains multi-layer user memory through internal stages for constraint extraction, memory fusion, and verification-gated updates; the Retrieval Agent performs hybrid vector--graph retrieval (GraphRAG); the Generation Agent produces evidence-referenced recommendations and explanations; and the Validation Agent applies multi-tier verification and targeted remediation. Together, these agents provide an auditable and reliable workflow for end-to-end housing consultation. We evaluate HabitatAgent on 100 real user consultation scenarios (300 multi-turn question--answer pairs) under an end-to-end correctness protocol. A strong single-stage baseline (Dense+Rerank) achieves 75% accuracy, while HabitatAgent reaches 95%.