RecSys Study: Simple Models Outperform Transformers on Multiple Benchmarks
_reachsumit · x · 2026-08-21
A paper accepted at RecSys 2026 questions whether widely used sequential recommendation benchmarks actually require higher-order sequence modeling. The study employs two simple, recency-weighted pairwise probe models: Sequential Rules (SeqRules) and Probabilistic Collaborative Transition Model (PCTM). Results show that on several Amazon datasets and MovieLens-1M, these simple probes outperform Transformer-based recommenders like eSASRec and SASRec by up to 38%, only trailing on MovieLens-20M. This suggests that many popular benchmarks may be poorly suited for measuring gains from higher-order sequence modeling.
More from Research
- Pretraining Potential: Coding Agents and the Compute Bottleneck — zeeshanp_ · 2026-08-21
- Study reveals why algorithms misalign: users engage more with misaligned content — msbernst · 2026-08-21
- PNAS Study: Social Algorithms Prioritize Content Clashing with Your Values — msbernst · 2026-08-21
- OneCanvas: Efficient 3D Scene Understanding via Panoramic Reprojection — rsasaki0109 · 2026-08-21
- Visualization Maps the Singularity from GPT-2 to Today — zakelfassi · 2026-08-21
- When Agents Collide: Why Software Engineering Still Holds the Keys to AGI's Future — maier_ak · 2026-08-21