Discover the AI behind what you watch, buy, and read. This category explores Recommender Systems, diving into the machine learning algorithms that drive highly personalized digital experiences 🛍️. From collaborative filtering to deep sequence modeling, explore the research that helps algorithms understand human behavior and connect users with exactly what they are looking for.
Most real recommendation catalogs are lopsided by design. A small number of items collect the overwhelming majority of clicks, purchases, or plays, while a much larger number sit with only a handful of interactions each. Collaborative filtering, the workhorse behind…
CHAKG Lets Popular Items Rescue Long Tail Recommendations Read More »
A criterion-wise local differential privacy framework perturbs each rating dimension on-device before it reaches the server — here’s what that privacy costs in recommendation accuracy.
What Happens When You Add Differential Privacy to a Multi-Criteria Recommender System Read More »
Federated recommenders drift apart when clients simply average parameters. AlignFedRec shares item similarity matrices instead, aligning structure rather than embeddings.
Why Federated Recommenders Break and How Structural Alignment Fixes Them Read More »
Researchers from Jilin University and the University of Technology Sydney built a federated continual recommendation framework that solves two problems nobody had tackled together: forgetting who you used to be, and not knowing enough about people like you — all…
FCUCR: Federated Continual Recommendation That Remembers You Without Storing Your Data Read More »
Engineers at Alibaba’s Xianyu platform built three interlocking improvements on top of their prior unified relevance-CTR model, ultimately achieving a 1.39% lift in per-capita orders and a 3.18% GMV increase in live A/B testing.
How a team of researchers at Kuaishou discovered that the biggest bottleneck in recommendation inference wasn’t model size or hardware—it was a mathematical redundancy hiding in plain sight, and how they fixed it without retraining a single parameter.
MaRI: How Kuaishou Solved the Hidden Redundancy Problem Plaguing Recommendation Models Read More »
Researchers from UW and Cornell prove that standard learning dynamics in competitive ML markets converge to disastrous equilibria where platforms serve only their existing fans, and introduce a knowledge-distillation-inspired fix that requires no retraining.