LLM-Based Re-Ranking for Real Estate Search
Information Retrieval
Summary
The authors from QuintoAndar Group developed a system that improves how people search for homes by using a large language model to better understand conversations with users. Instead of relying on rigid search filters, their system ranks home listings based on the detailed and changing needs expressed in user chats. They created a big dataset combining real and synthetic queries to test their method. Their evaluations showed that adding conversational understanding led to more people clicking on listings and scheduling visits, proving the approach helps match users with better housing options.
Authors
Nkateko Ntimane, Rafel Guedes, Tiago Cunha, Pedro Nogueira
Abstract
QuintoAndar Group operates the leading housing marketplace in Latin America for both rentals and sales. The platform replaces traditionally paper-heavy workflows with a fully digital experience, making housing transactions faster and more accessible to tenants, buyers, and landlords in the region. Finding the ideal home in such a vast catalog is inherently difficult. At the same time, the widespread adoption of conversational assistants is reshaping user expectations: people increasingly want to express their needs through open, multi-turn dialog rather than rigid filter menus and faceted search. This shift is particularly pronounced in housing, where intent is multi-dimensional, context-dependent, and rarely reducible to a small set of structured constraints. To meet these expectations, we propose a Large Language Model (LLM) based re-ranker that augments a conversational recommendation system by reordering retrieved candidates according to the nuanced, context-rich intent expressed across the user's conversation. We additionally construct a large-scale offline evaluation dataset for conversational real-estate search, containing 960,000 query-item pairs constructed from both synthetic and production queries and annotated using an LLM-as-a-Judge framework with human validation. We validate our approach both offline, on this proprietary dataset, and online, through a production A/B test. Both evaluations show consistent improvements in ranking quality, including a statistically significant increase in production of +5.3% in click-through rate and +4.8% in scheduled visits, demonstrating the value of integrating conversational context into housing recommendations.