Method
Listwise ranking
Optimize order of the whole list (ListMLE, LambdaRank).
10 cases from 9 companies apply Listwise ranking. 4 were checked in an online A/B test; 1 report results in production.
Typical tasks: Rank search results, Retrieve and rank from millions of items, Recommend from what the user just did, Label data with LLMs and 1 more. Companies: OLX, Avito, Algolia, LinkedIn, Mercari, Nextdoor and 3 more. Often combined with: Gradient boosting, A/B test, Feature store, Deep & Cross Network, Blending and 25 more.
| Neural Ranking for Recommendations | classifieds | 2026-05 | |
| How Vinted Serves Personalised Search Autocomplete | marketplace | 2026-04 | |
| How we scaled ranking with Learn-to-Rank | e-commerce | 2026-02 | |
| Building the next generation of job search at LinkedIn | marketplace | 2025-05 | |
| From RankNet to LambdaMART: Leveraging XGBoost for Enhanced Ranking Models | classifieds | 2025-02 | |
| Building a Listwise Ranking TF recommender: A step-by-step guide | food-delivery | 2023-08 | |
| Position Bias in Recommender Systems | e-commerce | 2023-05 | |
The Journey to Machine-Learned Re-ranking![]() | ![]() | marketplace | 2023-01 |
| Learning-To-Rank: Sorting Search Results | classifieds | 2022-06 | |
| Modernizing Nextdoor Search Stack — Part 2 | social | 2022-05 |
