LLM-Based Re-Ranking for Real Estate Search¶
Fidelity: core mechanism. Paper results, local public-data measurements, and omissions are separate.
Paper information¶
| Field | Details |
|---|---|
| Paper | arXiv 2607.14835 |
| Organization | QuintoAndar |
| First public release | 2026-07-16 (arXiv v1) |
| Authors' code | The paper did not provide official or author code when checked on 2026-08-24. |
| Adapter / method | real-estate-rerank |
| Local reproduction code | src/auto_research/reproductions/real_estate_rerank/ |
Original paper¶
Motivation and method¶
Pass conversational requirements, structured listing attributes, text descriptions, and candidate-set statistics to an LLM reranker. Cross-candidate comparison complements local relevance from vector retrieval.
flowchart LR
A["Conversation and user profile"] --> B["Candidate-set LLM reranker"] --> C["Listing order"]Key figure from the paper¶
Figure 4: Overall workflow, key stages, and data flow. From the original paper; copyright remains with the authors.
Core equations¶
Paper results¶
The main production A/B evidence is CTR +5.30%. Offline tables and online metrics use private or paper-specific protocols. The private paper protocol is not directly comparable with local MovieLens measurements.
Local reproduction¶
Local comparison: The baseline is the shared transition + content scorer. The treatment adds only real-estate-rerank on the same users, items, full-catalog candidates, and seed.
MovieLens-100K, 220 users and 360 items, seed 42: NDCG@10 0.0540 → 0.0496 (-8.21%); Hit@10 0.1091 → 0.1000. Validation selects mixture weights only; the test set is not used for tuning.
See the fixed metrics.
Reproduction boundaries¶
The core mechanism runs on MovieLens-100K with the same split, fixed full-catalog candidates, and seed. Private features, the production foundation model, and online traffic are unavailable. The implementation has independent model state and scoring, rather than mapping a paper name to a shared weighted formula. Omitted parts are in the adapter's omitted_core_components. Local changes and paper A/B gains are reported separately.