dorsal/arxiv
View SchemaScalable Sequential Recommendation under Latency and Memory Constraints
| Authors | Adithya Parthasarathy, Aswathnarayan Muthukrishnan Kirubakaran, Vinoth Punniyamoorthy, Nachiappan Chockalingam, Lokesh Butra, Kabilan Kannan, Abhirup Mazumder, Sumit Saha |
|---|---|
| Categories | |
| ArXiv ID | 2601.08360vv2 |
| URL | https://arxiv.org/abs/2601.08360 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Sequential recommender systems must model long-range user behavior while operating under strict memory and latency constraints. Transformer-based approaches achieve strong accuracy but suffer from quadratic attention complexity, forcing aggressive truncation of user histories and limiting their practicality for long-horizon modeling. This paper presents HoloMambaRec, a lightweight sequential recommendation architecture that combines holographic reduced representations for attribute-aware embedding with a selective state space encoder for linear-time sequence processing. Item and attribute information are bound using circular convolution, preserving embedding dimensionality while encoding structured metadata. A shallow selective state space backbone, inspired by recent Mamba-style models, enables efficient training and constant-time recurrent inference. Experiments on Amazon Beauty and MovieLens-1M datasets demonstrate that HoloMambaRec consistently outperforms SASRec and achieves competitive performance with GRU4Rec under a constrained 10-epoch training budget, while maintaining substantially lower memory complexity. The design further incorporates forward-compatible mechanisms for temporal bundling and inference-time compression, positioning HoloMambaRec as a practical and extensible alternative for scalable, metadata-aware sequential recommendation.
{
"annotation_id": "04fb83ce-e1f8-45e7-a74a-83d5ed86e641",
"date_created": "2026-02-17T05:53:16.140000Z",
"date_modified": "2026-02-17T05:53:16.140000Z",
"file_hash": "28b4b0acf1545b636074d6bfae9786952f60180aca5a936013770b9749136905",
"private": false,
"record": {
"abstract": "Sequential recommender systems must model long-range user behavior while operating under strict memory and latency constraints. Transformer-based approaches achieve strong accuracy but suffer from quadratic attention complexity, forcing aggressive truncation of user histories and limiting their practicality for long-horizon modeling. This paper presents HoloMambaRec, a lightweight sequential recommendation architecture that combines holographic reduced representations for attribute-aware embedding with a selective state space encoder for linear-time sequence processing. Item and attribute information are bound using circular convolution, preserving embedding dimensionality while encoding structured metadata. A shallow selective state space backbone, inspired by recent Mamba-style models, enables efficient training and constant-time recurrent inference. Experiments on Amazon Beauty and MovieLens-1M datasets demonstrate that HoloMambaRec consistently outperforms SASRec and achieves competitive performance with GRU4Rec under a constrained 10-epoch training budget, while maintaining substantially lower memory complexity. The design further incorporates forward-compatible mechanisms for temporal bundling and inference-time compression, positioning HoloMambaRec as a practical and extensible alternative for scalable, metadata-aware sequential recommendation.",
"arxiv_id": "2601.08360",
"authors": [
"Adithya Parthasarathy",
"Aswathnarayan Muthukrishnan Kirubakaran",
"Vinoth Punniyamoorthy",
"Nachiappan Chockalingam",
"Lokesh Butra",
"Kabilan Kannan",
"Abhirup Mazumder",
"Sumit Saha"
],
"categories": [
"cs.IR",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Scalable Sequential Recommendation under Latency and Memory Constraints",
"url": "https://arxiv.org/abs/2601.08360",
"version": "v2"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "201dbde7-8688-4d29-b55f-308887358ae9",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}