LLM re-ranker boosts real estate search with conversational context

Researchers have developed a new Large Language Model (LLM) based re-ranker to improve real estate search on the QuintoAndar Group platform. This system augments conversational recommendation systems by reordering property listings based on the nuanced intent expressed in user conversations, moving beyond traditional filter menus. An extensive offline evaluation dataset was created, and A/B testing in production showed a significant increase in click-through rates and scheduled visits, demonstrating the effectiveness of incorporating conversational context into housing recommendations. AI

IMPACT Enhances LLM application in specialized search domains, improving user experience and conversion rates.

RANK_REASON The item is a research paper detailing a new method for improving search using LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.IR (Information Retrieval) →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Latin America
• LLM-as-a-Judge
• LLM-Based Re-Ranking for Real Estate Search
• QuintoAndar Group
• ScienceCast
• Tiago Cunha

• product
• paper

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →