Pay Attention to Sequence Split: Uncovering the Impacts of Sub-Sequence Splitting on Sequential Recommendation Models

2026-04-06

AI Summary

Subsequence segmentation (SSS) is widely adopted in sequential recommendation, yet its impact on model evaluation and the conditions under which it is effective remain underexplored. This work is the first to reveal that SSS can distort evaluation outcomes. Through ablation studies, distributional analyses, and multidimensional comparisons, we demonstrate that the performance gains attributed to SSS stem primarily from its ability to improve data distribution balance and enhance target item coverage—and that these benefits manifest only under specific combinations of target sampling strategies and loss functions. Notably, removing SSS leads to significant performance degradation in multiple state-of-the-art models. Based on these findings, we offer practical recommendations to avoid evaluation bias and release our code and data to foster a more reproducible and equitable evaluation paradigm.

Abstract

Sub-sequence splitting (SSS) has been demonstrated as an effective approach to mitigate data sparsity in sequential recommendation (SR) by splitting a raw user interaction sequence into multiple sub-sequences. Previous studies have demonstrated its ability to enhance the performance of SR models significantly. However, in this work, we discover that \textbf{(i). SSS may interfere with the evaluation of the model's actual performance.} We observed that many recent state-of-the-art SR models employ SSS during the data reading stage (not mentioned in the papers). When we removed this operation, performance significantly declined, even falling below that of earlier classical SR models. The varying improvements achieved by SSS and different splitting methods across different models prompt us to analyze further when SSS proves effective. We find that \textbf{(ii). SSS demonstrates strong capabilities only when specific splitting methods, target strategies, and loss functions are used together.} Inappropriate combinations may even harm performance. Furthermore, we analyze why sub-sequence splitting yields such remarkable performance gains and find that \textbf{(iii). it evens out the distribution of training data while increasing the likelihood that different items are targeted.} Finally, we provide suggestions for overcoming SSS interference, along with a discussion on data augmentation methods and future directions. We hope this work will prompt the broader community to re-examine the impact of data splitting on SR and promote fairer, more rigorous model evaluation. All analysis code and data will be made available upon acceptance. We provide a simple, anonymous implementation at https://github.com/KingGugu/SSS4SR.

Problem

• Sub-sequence Splitting
• Sequential Recommendation
• Model Evaluation
• Data Sparsity
• Performance Interference

Innovation

• Sub-Sequence Splitting
• Sequential Recommendation
• Data Sparsity
• Model Evaluation
• Data Augmentation