Agentic Recommender System with Hierarchical Belief-State Memory

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Xiang, Zhou, Yuhang, Wu, Yifan, Zhao, Zhuokai, Lin, Siyu, Huang, Lei, Zhong, Qianqian, Zhang, Lizhu, Zhang, Benyu, Fan, Xiangjun, Yan, Hong
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910223337259008
author Shen, Xiang
Zhou, Yuhang
Wu, Yifan
Zhao, Zhuokai
Lin, Siyu
Huang, Lei
Zhong, Qianqian
Zhang, Lizhu
Zhang, Benyu
Fan, Xiangjun
Yan, Hong
author_facet Shen, Xiang
Zhou, Yuhang
Wu, Yifan
Zhao, Zhuokai
Lin, Siyu
Huang, Lei
Zhong, Qianqian
Zhang, Lizhu
Zhang, Benyu
Fan, Xiangjun
Yan, Hong
contents Memory-augmented LLM agents have advanced personalized recommendation, yet existing approaches universally adopt flat memory representations that conflate ephemeral signals with stable preferences, and none provides a complete lifecycle governing how memory should evolve. We propose MARS (Memory-Augmented Agentic Recommender System), a framework that treats recommendation as a partially observable problem and maintains a structured belief state that progressively abstracts noisy behavioral observations into a compact estimate of user preferences. MARS organizes this belief state into three tiers: event memory buffers raw signals, preference memory maintains fine-grained mutable chunks with explicit strength and evidence tracking, and profile memory distills all preferences into a coherent natural language narrative. A complete lifecycle of six operations -- extraction, reinforcement, weakening, consolidation, forgetting, and resynthesis -- is adaptively scheduled by an LLM-based planner rather than fixed-interval heuristics. Experiments on four InstructRec benchmark domains show that MARS achieves state-of-the-art performance with average improvements of 26.4% in HR@1 and 10.3% in NDCG@10 over the strongest baselines with further gains from agentic scheduling in evolving settings.
format Preprint
id arxiv_https___arxiv_org_abs_2605_14401
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Agentic Recommender System with Hierarchical Belief-State Memory
Shen, Xiang
Zhou, Yuhang
Wu, Yifan
Zhao, Zhuokai
Lin, Siyu
Huang, Lei
Zhong, Qianqian
Zhang, Lizhu
Zhang, Benyu
Fan, Xiangjun
Yan, Hong
Computation and Language
Artificial Intelligence
Memory-augmented LLM agents have advanced personalized recommendation, yet existing approaches universally adopt flat memory representations that conflate ephemeral signals with stable preferences, and none provides a complete lifecycle governing how memory should evolve. We propose MARS (Memory-Augmented Agentic Recommender System), a framework that treats recommendation as a partially observable problem and maintains a structured belief state that progressively abstracts noisy behavioral observations into a compact estimate of user preferences. MARS organizes this belief state into three tiers: event memory buffers raw signals, preference memory maintains fine-grained mutable chunks with explicit strength and evidence tracking, and profile memory distills all preferences into a coherent natural language narrative. A complete lifecycle of six operations -- extraction, reinforcement, weakening, consolidation, forgetting, and resynthesis -- is adaptively scheduled by an LLM-based planner rather than fixed-interval heuristics. Experiments on four InstructRec benchmark domains show that MARS achieves state-of-the-art performance with average improvements of 26.4% in HR@1 and 10.3% in NDCG@10 over the strongest baselines with further gains from agentic scheduling in evolving settings.
title Agentic Recommender System with Hierarchical Belief-State Memory
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2605.14401