SPADE: new metric measures serendipity as distance from per-user popularity-similarity Pareto front
_reachsumit · x · 2026-09-28
An arXiv paper introduces SPADE (Serendipitous Pareto Distance Evaluation), an offline metric for serendipity in recommender systems.
- Existing beyond-accuracy metrics tend to isolate either historical similarity or global popularity
- SPADE maps items into a 2D popularity-similarity space and computes a per-user Pareto frontier of maximally popular, history-similar items
- The serendipity score averages the minimum Euclidean distance from that frontier, over correctly recommended test items only
- Evaluated on 5 datasets against 5 baselines, it prevents gaming via irrelevant or non-personalized recommendations and reliably isolates genuine serendipitous discoveries
More from Research
- Stanford-led paper traces 'bigger is better' to data-induced competition for neurons — niloofar_mire · 2026-10-24
- Frequency drives conventionalization: higher-exposure communities conventionalize more words — abenitezburraco · 2026-10-24
- GeoWind2Plan: seconds-scale 3D urban wind prediction for energy-efficient drones — AnimaAnandkumar · 2026-10-24
- Flowing water and lava come to 3D Gaussian Splatting scenes — jonstephens85 · 2026-10-24
- Immorta Bio's AI-designed 9-amino-acid peptide SenoMune-0257 targets senescent cells — DavidmComfort · 2026-10-24
- France rebuilt its tax code as MLang, a formally verified open-source compiler — fkasummer · 2026-10-24