AI summaryⓘ
The authors address a problem in vacation rental marketplaces where most user interactions focus on a few popular properties, leaving many less popular ones without good recommendations. They propose using a large language model (LLM) to generate search queries based only on property details, complementing the usual method that relies on user behavior data. Their approach combines both methods to recommend a wider range of properties, especially benefiting those with little user interaction, without hurting recommendations for popular listings. They also show that this fusion approach narrows the performance gap between smaller, self-hosted LLMs and larger, API-based models, making it practical to use smaller models at large scale.
vacation rental marketplacesitem-based k-nearest neighbors (IBKNN)large language model (LLM)candidate generationcollaborative filteringsemantic queriesapproximate nearest neighborslearning-to-ranklong tail problemUnion fusion strategy
Authors
Syed Mohammed Arshad Zaidi, Eric Rincon, Shayan Hassantabar
Abstract
Vacation rental marketplaces face a structural imbalance on the supply side: a small fraction of properties receive most user interactions, while the long tail of new, niche, and seasonal listings generates too little behavioral signal for collaborative filtering to serve effectively. At Vrbo, item-based k-nearest neighbors (IBKNN) is a core candidate generation channel, but leaves tens of thousands of properties with no candidates and produces weak neighborhoods for sparsely interacted ones. We present a training-free, LLM-based candidate generation pipeline that complements IBKNN using static property metadata alone. An off-the-shelf LLM synthesizes diverse semantic queries per property, a pre-trained text encoder embeds them, and an approximate nearest-neighbor index retrieves candidates from an 11.7M-property catalog. A Union fusion strategy merges these with IBKNN while preserving the behavioral channel's ordering, guaranteeing no degradation on well-served properties, and a downstream learning-to-rank model re-scores the fused pool. Evaluated on 1.6M focal properties, the system extends candidate coverage to tens of thousands of properties IBKNN cannot reach, delivers its largest gains on the long-tail segment where behavioral methods are weakest, and matches or beats IBKNN at every K on shared properties. A downstream learning-to-rank stage further lifts the fused pool, yielding a complete candidate generation and re-ranking stack that serves the long tail without regressing well-served properties. We additionally show that Union fusion collapses the recall gap between a 3B open-weights LLM and frontier API-based models from 27-46% to under 1%, supporting self-hosted small-model deployment at marketplace catalog scale.