# A Tale of Two Graphs: Freezing and Denoising Graph Structures for Multimodal Recommendation

arXiv: https://arxiv.org/abs/2211.06924 · HTML: https://mlsanity.com/paper/2211.06924/

## Key facts

- Cited by 54 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Nov 2022.

## Built on by (most cited first, 30 of 54)

- [Multimodal Recommender Systems: A Survey](https://mlsanity.com/paper/2302.03883/) (2023)
- [AlignRec: Aligning and Training in Multimodal Recommendations](https://mlsanity.com/paper/2403.12384/) (2024)
- [Rec-GPT4V: Multimodal Recommendation with Large Vision-Language Models](https://mlsanity.com/paper/2402.08670/) (2024)
- [Graph World Model](https://mlsanity.com/paper/2507.10539/) (2025)
- [Multimodality Invariant Learning for Multimedia-Based New Item Recommendation](https://mlsanity.com/paper/2405.15783/) (2024)
- [GUME: Graphs and User Modalities Enhancement for Long-Tail Multimodal Recommendation](https://mlsanity.com/paper/2407.12338/) (2024)
- [A Survey on Large Language Models in Multimodal Recommender Systems](https://mlsanity.com/paper/2505.09777/) (2025)
- [Mirror Gradient: Towards Robust Multimodal Recommender Systems via Exploring Flat Local Minima](https://mlsanity.com/paper/2402.11262/) (2024)
- [Structured Spectral Reasoning for Frequency-Adaptive Multimodal Recommendation](https://mlsanity.com/paper/2512.01372/) (2025)
- [Modality-Balanced Learning for Multimedia Recommendation](https://mlsanity.com/paper/2408.06360/) (2024)
- [LinkedOut: Linking World Knowledge Representation Out of Video LLM for Next-Generation Video Recommendation](https://mlsanity.com/paper/2512.16891/) (2025)
- [A Unified Graph Transformer for Overcoming Isolations in Multi-modal Recommendation](https://mlsanity.com/paper/2407.19886/) (2024)
- [Disentangling and Generating Modalities for Recommendation in Missing Modality Scenarios](https://mlsanity.com/paper/2504.16352/) (2025)
- [RecGOAT: Graph Optimal Adaptive Transport for LLM-Enhanced Multimodal Recommendation with Dual Semantic Alignment](https://mlsanity.com/paper/2602.00682/) (2026)
- [ID Embedding as Subtle Features of Content and Structure for Multimodal Recommendation](https://mlsanity.com/paper/2311.05956/) (2023)
- [Do Recommender Systems Really Leverage Multimodal Content? A Comprehensive Analysis on Multimodal Representations for Recommendation](https://mlsanity.com/paper/2508.04571/) (2025)
- [Training-free Adjustable Polynomial Graph Filtering for Ultra-fast Multimodal Recommendation](https://mlsanity.com/paper/2503.04406/) (2025)
- [RGL: A Graph-Centric, Modular Framework for Efficient Retrieval-Augmented Generation on Graphs](https://mlsanity.com/paper/2503.19314/) (2025)
- [MMHCL: Multi-Modal Hypergraph Contrastive Learning for Recommendation](https://mlsanity.com/paper/2504.16576/) (2025)
- [Semantic Item Graph Enhancement for Multimodal Recommendation](https://mlsanity.com/paper/2508.06154/) (2025)
- [BiVRec: Bidirectional View-based Multimodal Sequential Recommendation](https://mlsanity.com/paper/2402.17334/) (2024)
- [Dreaming User Multimodal Representation Guided by The Platonic Representation Hypothesis for Micro-Video Recommendation](https://mlsanity.com/paper/2410.03538/) (2024)
- [Echoless Label-Based Pre-computation for Memory-Efficient Heterogeneous Graph Learning](https://mlsanity.com/paper/2511.11081/) (2025)
- [Benchmarking Multimodal Large Language Models for Missing Modality Completion in Product Catalogues](https://mlsanity.com/paper/2601.19750/) (2026)
- [MoToRec: Sparse-Regularized Multimodal Tokenization for Cold-Start Recommendation](https://mlsanity.com/paper/2602.11062/) (2026)
- [LAC: Graph Contrastive Learning with Learnable Augmentation in Continuous Space](https://mlsanity.com/paper/2410.15355/) (2024)
- [Bridging Domain Gaps between Pretrained Multimodal Models and Recommendations](https://mlsanity.com/paper/2502.15542/) (2025)
- [EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation](https://mlsanity.com/paper/2508.16170/) (2025)
- [Multimodal Representation-disentangled Information Bottleneck for Multimodal Recommendation](https://mlsanity.com/paper/2509.20225/) (2025)
- [Causal Inspired Multi Modal Recommendation](https://mlsanity.com/paper/2510.12325/) (2025)

## Abstract (arXiv)

Multimodal recommender systems utilizing multimodal features (e.g., images and textual descriptions) typically show better recommendation accuracy than general recommendation models based solely on user-item interactions. Generally, prior work fuses multimodal features into item ID embeddings to enrich item representations, thus failing to capture the latent semantic item-item structures. In this context, LATTICE proposes to learn the latent structure between items explicitly and achieves state-of-the-art performance for multimodal recommendations. However, we argue the latent graph structure learning of LATTICE is both inefficient and unnecessary. Experimentally, we demonstrate that freezing its item-item structure before training can also achieve competitive performance. Based on this finding, we propose a simple yet effective model, dubbed as FREEDOM, that FREEzes the item-item graph and DenOises the user-item interaction graph simultaneously for Multimodal recommendation. Theoretically, we examine the design of FREEDOM through a graph spectral perspective and demonstrate that it possesses a tighter upper bound on the graph spectrum. In denoising the user-item interaction graph, we devise a degree-sensitive edge pruning method, which rejects possibly noisy edges with a high probability when sampling the graph. We evaluate the proposed model on three real-world datasets and show that FREEDOM can significantly outperform current strongest baselines. Compared with LATTICE, FREEDOM achieves an average improvement of 19.07% in recommendation accuracy while reducing its memory cost up to 6× on large graphs. The source code is available at: https://github.com/enoche/FREEDOM.