Papers › Lightweight representation learning for efficient and scalable recommendation

Lightweight representation learning for efficient and scalable recommendation

4 Jan 2021arXiv:2101.00870archive 2025-07-28

Olivier Koch, Amine Benhalloum, Guillaume Genthial, Denis Kuzin, Dmitry Parfenchik

Over the past decades, recommendation has become a critical component of many online services such as media streaming and e-commerce. Recent advances in algorithms, evaluation methods and datasets have led to continuous improvements of the state-of-the-art. However, much work remains to be done to make these methods scale to the size of the internet. Online advertising offers a unique testbed for recommendation at scale. Every day, billions of users interact with millions of products in real-time. Systems addressing this scenario must work reliably at scale. We propose an efficient model (LED, for Lightweight Encoder-Decoder) reaching a new trade-off between complexity, scale and performance. Specifically, we show that combining large-scale matrix factorization with lightweight embedding fine-tuning unlocks state-of-the-art performance at scale. We further provide the detailed description of a system architecture and demonstrate its operation over two months at the scale of the internet. Our design allows serving billions of users across hundreds of millions of items in a few milliseconds using standard hardware.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

criteo/deepr officialmentioned in papertf report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderRecommendation SystemsRepresentation Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Recommendation Systems MovieLens 20M LED Recall@20 0.375 #14 of 18 Archive leaderboard report
Recommendation Systems MovieLens 20M LED Recall@50 0.516 #14 of 18 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections