MemRec

Collaborative Memory-Augmented Agentic Recommender System

🎮

Try Demo

Experience MemRec in an interactive simulation! Step through our demo to observe how collaborative memory augments agentic recommendations.

Launch Demo →
🔬

Agentic RS Hub

Explore our curated collection of cutting-edge research on agentic reasoning, memory mechanisms, and LLMs in recommender systems.

Explore Hub →

📖 About MemRec

The evolution of recommender systems has shifted preference storage from rating matrices and dense embeddings to semantic memory in the agentic era. Yet existing agents rely on isolated memory, overlooking crucial collaborative signals. Bridging this gap is hindered by the dual challenges of distilling vast graph contexts without overwhelming reasoning agents with cognitive load, and evolving the collaborative memory efficiently without incurring prohibitive computational costs.

To address this, we propose MemRec, a framework that architecturally decouples reasoning from memory management to enable efficient collaborative augmentation. MemRec introduces a dedicated, cost-effective LMMem to manage a dynamic collaborative memory graph, serving synthesized, high-signal context to a downstream LLMRec. The framework operates via a practical pipeline featuring efficient retrieval and cost-effective asynchronous graph propagation that evolves memory in the background.

🌟 Key Features

  • Architectural Decoupling: Separates reasoning from memory management for efficiency
  • Collaborative Memory Graph: Captures and evolves collaborative signals dynamically
  • Cost-Effective Design: Dedicated LMMem reduces computational burden
  • State-of-the-Art Performance: Validated on four benchmarks
  • Flexible Deployment: Supports diverse models including local open-source options
  • Privacy-Preserving: Balances performance with privacy through architectural design

Extensive experiments on four benchmarks demonstrate that MemRec achieves state-of-the-art performance. Furthermore, architectural analysis confirms its flexibility, establishing a new Pareto frontier that balances reasoning quality, cost, and privacy through support for diverse deployments.

📋 Citation

If you find MemRec useful in your research, please consider citing our paper: 

@article{chen2026memrec,
  title   = {MemRec: Collaborative Memory-Augmented Agentic Recommender System},
  author  = {Chen, Weixin and Zhao, Yuhan and Huang, Jingyuan and Ye, Zihe and 
             Ju, Clark Mingxuan and Zhao, Tong and Shah, Neil and Chen, Li and 
             Zhang, Yongfeng},
  year    = 2026,
  journal = {arXiv preprint arXiv:2601.08816},
  url     = {https://arxiv.org/abs/2601.08816}
}
📄 Read Paper 💻 GitHub Repo