Is Sliding Window All You Need? An Open Framework for Long-Sequence Recommendation

TL;DR

This paper introduces an open framework for long-sequence recommendation using sliding windows, demonstrating its practicality and effectiveness with new techniques like a runtime-aware ablation and a novel embedding layer.

Contribution

It provides a complete, open-source pipeline for long-sequence training, including new methods that improve efficiency and scalability on commodity hardware.

Findings

Achieved up to +6.04% MRR and +6.34% Recall@10 on Retailrocket

Demonstrated reliable training on modest university clusters

Introduced a k-shift embedding layer enabling million-scale vocabularies

Abstract

Long interaction histories are central to modern recommender systems, yet training with long sequences is often dismissed as impractical under realistic memory and latency budgets. This work demonstrates that it is not only practical but also effective-at academic scale. We release a complete, end-to-end framework that implements industrial-style long-sequence training with sliding windows, including all data processing, training, and evaluation scripts. Beyond reproducing prior gains, we contribute two capabilities missing from earlier reports: (i) a runtime-aware ablation study that quantifies the accuracy-compute frontier across windowing regimes and strides, and (ii) a novel k-shift embedding layer that enables million-scale vocabularies on commodity GPUs with negligible accuracy loss. Our implementation trains reliably on modest university clusters while delivering competitive retrieval quality (e.g., up to +6.04% MRR and +6.34% Recall@10 on Retailrocket) with $\sim 4 \times $ training-time overheads. By packaging a robust pipeline, reporting training time costs, and introducing an embedding mechanism tailored for low-resource settings, we transform long-sequence training from a closed, industrial technique into a practical, open, and extensible methodology for the community.