# Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation

arXiv: https://arxiv.org/abs/2607.07108 · HTML: https://mlsanity.com/paper/2607.07108/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Agentic recommender systems, one of 16 directions in LLM agents and tool use.
- Builds on 18 of its 22 references that are in the map.
- First posted on arXiv 8 Jul 2026.

## Placed in

- LLM agents and tool use / [Agentic recommender systems](https://mlsanity.com/agents/agentic-recommender-systems/)

## Builds on (most cited first, 18 of 18)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [A Survey on the Memory Mechanism of Large Language Model based Agents](https://mlsanity.com/paper/2404.13501/) (2024)
- [TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation](https://mlsanity.com/paper/2305.00447/) (2023)
- [Large Language Models are Zero-Shot Rankers for Recommender Systems](https://mlsanity.com/paper/2305.08845/) (2023)
- [GME: Improving Universal Multimodal Retrieval by Multimodal LLMs](https://mlsanity.com/paper/2412.16855/) (2024)
- [From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review](https://mlsanity.com/paper/2504.19678/) (2025)
- [AgentCF: Collaborative Learning with Autonomous Language Agents for Recommender Systems](https://mlsanity.com/paper/2310.09233/) (2023)
- [Leveraging Large Language Models for Sequential Recommendation](https://mlsanity.com/paper/2309.09261/) (2023)
- [AgentCF++: Memory-enhanced LLM-based Agents for Popularity-aware Cross-domain Recommendations](https://mlsanity.com/paper/2502.13843/) (2025)
- [Retrieval Augmented Generation with Collaborative Filtering for Personalized Text Generation](https://mlsanity.com/paper/2504.05731/) (2025)
- [A Survey of Foundation Model-Powered Recommender Systems: From Feature-Based, Generative to Agentic Paradigms](https://mlsanity.com/paper/2504.16420/) (2025)
- [A Collaborative Multi-Agent Approach to Retrieval-Augmented Generation Across Diverse Data](https://mlsanity.com/paper/2412.05838/) (2024)
- [Dual Conditional Diffusion Models for Sequential Recommendation](https://mlsanity.com/paper/2410.21967/) (2024)
- [Dual Contrastive Transformer for Hierarchical Preference Modeling in Sequential Recommendation](https://mlsanity.com/paper/2410.22790/) (2024)
- [MemWeaver: Weaving Hybrid Memories for Traceable Long-Horizon Agentic Reasoning](https://mlsanity.com/paper/2601.18204/) (2026)
- [Modeling Temporal Positive and Negative Excitation for Sequential Recommendation](https://mlsanity.com/paper/2410.22013/) (2024)
- [Factorized Latent Reasoning for LLM-based Recommendation](https://mlsanity.com/paper/2604.26760/) (2026)

## Built on by (most cited first, 2 of 2)

- [KREL: Automatic Medical Coding via Knowledge-Guided Reasoning over Clinical Evidence with LLMs](https://mlsanity.com/paper/2608.20887/) (2026)
- [AUSO: Action-Level Unified Skill Optimization from Internalization to Utilization](https://mlsanity.com/paper/2608.21292/) (2026)

## Abstract (arXiv)

Large language model (LLM)-based agentic recommender systems show promise in modeling user preferences through natural-language reasoning, yet they remain limited by text-centric inputs and coarse-grained memory updates, making agents prone to missing visual evidence, semantic noise, and preference drift. To address these limitations, we propose MMEACR, a Multimodal Memory-Enhanced Agent Collaboration framework for recommendation. MMEACR introduces a dual-track memory architecture that separates interpretable agent reasoning from fine-grained multimodal matching. In the reasoning track, collaborative User and Item Memory Agents maintain persistent multimodal memories and update them through an attribute-guided reinforcement-and-reflection mechanism. In the matching track, a decoupled multi-modal embedding memory is built from raw interaction narratives and item images to preserve detailed cross-modal signals beyond structured memory updates. The two tracks are integrated through weighted Reciprocal Rank Fusion to produce robust and interpretable rankings. Experiments on three real-world domains show that MMEACR achieves strong overall performance against competitive LLM-based and agent-based baselines, with notable gains in visually grounded recommendation scenarios.