Skip to content
Book Open access

Information-Aware Long Sequence Compression for Sequential Recommendation

Sep 2026 · Proceedings of the 20th ACM Conference on Recommender Systems · 0 citations · 22 references

TL;DR

RDSR, a Rate–Distortion–based Sequential Recommendation framework grounded in a task-oriented rate-utility view, combines fixed-capacity token selection with VIB-based latent compression to retain task-relevant information under explicit rate control, effectively reducing sequence length and computational overhead.

Abstract

Sequential recommendation (SR) aims to predict a user’s next interaction by modeling temporal dependencies in historical behavior sequences. However, modeling long sequences introduces two challenges: longer histories often include noisy interactions irrelevant to a user’s core interests, and increasing sequence length substantially raises computational cost while often degrading prediction accuracy due to noise accumulation. We present RDSR, a Rate–Distortion–based Sequential Recommendation framework grounded in a task-oriented rate-utility view. Instead of directly modeling full-length sequences, RDSR combines fixed-capacity token selection with VIB-based latent compression to retain task-relevant information under explicit rate control. This suppresses irrelevant interactions while preserving essential preference signals, effectively reducing sequence length and computational overhead. Extensive experiments show that RDSR improves the performance–efficiency trade-off across attention-, MLP-, and SSM-based SR backbones, with performance gains depending on backbone inductive bias. Our code and supplementary material are available at https://github.com/kangvic0615/Rate-Distortion-SR

Read PDF

Similar papers

Book Open access Sep 2026

DP-Rec: Towards Dynamic Patching for Efficient Long-Sequence Recommendation

Inspired by the recent success of Byte Latent Transformer, DP-Rec is proposed, a dynamic latent patching architecture for recommendation that scales effectively to long sequences and achieves a superior efficiency–accuracy trade-off over both non-compressed and fixed-size compression baselines.

Dwipam Katariya, T. Caputo, Akshat Shreemali et al. · 0 citations
Preprint Sep 2026

Closing the Long-Short View Gap in Sequential Recommendation without Cached History

This paper empirically identifies two structural flaws rooted in geometric properties and dataset sparsity, and proposes a novel two-stage framework to close the long-short-view performance gap and introduces a modified softmax to counter prefix position bias.

Ling-Feng Shi, Chengkai Huang, Li-Na Yao et al. · 0 citations
Preprint Sep 2026

Preference-Drift-Aware Subsequence Learning and Hierarchical Context Fusion for Long-Sequence Generative Recommendation

Long-sequence generative recommendation methods autoregressively model the user's interaction sequence to generate the next-item representation. Existing methods generally fall into two categories: efficient full-sequence modeling and target-aware context retrieval. Our experiments reveal that as the sequence length in...

Fei Li, Qing-Yun Gao, Jian-Zhe Zhao et al. · 0 citations
Preprint Sep 2026

MuSeR: Scalable Long-sequence Recommendation with Multi-interest Modeling

The contribution is a system-level integration that makes long-term, multi-interest, and multimodal modeling jointly deployable in a real-time production pipeline, together with the engineering practices required to sustain it.

Yong-Kang Fu, Bei-Ning Bao, Yu Jiang et al. · 0 citations
Book Open access Sep 2026

UniTraj: Cross-Domain Long-Sequence Modeling for Commercial Recommendation

UniTraj, a practical framework that extends sequence construction beyond the advertising domain by incorporating behaviors from content-consumption scenarios, forming unified commercial trajectories across domains and scenarios, is proposed and deployed in a large-scale online advertising system.

Xian Hu, Ming Yue, Zhi-Xiang Feng et al. · 0 citations
Book Open access Sep 2026

Interest Sequence for User Modeling in Industrial Short-Form Video Recommendation

Ultra-long user history modeling has been a highly effective approach in modern industrial recommendation systems, with most works heavily utilizing search-based methods and summarization based methods to map massive user interaction logs into latent user representation with algorithm designs to address the scaling cha...

Yuan-Zheng Lin, Diego Uribe Mora, Yuan Shao et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.