Anchored Alignment: Preventing Positional Collapse in Multimodal Recommender Systems

Yonghun Jeong; David Yoon Suk Kang; Yeon-Chang Lee

cs.IRMar 13, 2026

Anchored Alignment: Preventing Positional Collapse in Multimodal Recommender Systems

Authors: Yonghun Jeong, David Yoon Suk Kang, Yeon-Chang Lee

Organizations: UNIST, Ulsan, Korea · Chungbuk National University, Cheongju, Korea

Abstract

Multimodal recommender systems (MMRS) leverage images, text, and interaction signals to enrich item representations. However, recent alignment based MMRSs that enforce a unified embedding space often blur modality specific structures and exacerbate ID dominance. Therefore, we propose AnchorRec, a multimodal recommendation framework that performs indirect, anchor based alignment in a lightweight projection domain. By decoupling alignment from representation learning, AnchorRec preserves each modality's native structure while maintaining cross modal consistency and avoiding positional collapse. Experiments on four Amazon datasets show that AnchorRec achieves competitive top N recommendation accuracy, while qualitative analyses demonstrate improved multimodal expressiveness and coherence. The codebase of AnchorRec is available at https://github.com/hun9008/AnchorRec.

Figures & tables

Figure 1. t-SNE ( Maaten and Hinton, 2008 ) visualization of embeddings from AlignRec ( Liu et al., 2024 ) : the left panel shows the global distribution, and the right panel provides a zoomed-in local view revealing fine-grained local structure.

Figure 2. Overview of \ours , highlighting the anchor-based projection and its alignment losses as the core mechanism.

Figure 3. Comparison of alignment strategies: (a) \ours : anchor-based indirect alignment via a shared multimodal anchor ( $\boldsymbol{\mathbf{p}_{\mathrm{mm}}^{i}}$ ); (b) BM3 ( Zhou et al., 2023b ) : direct similarity-based alignment across modalities; and (c) AlignRec ( Liu et al., 2024 ) : direct contrastive alignment between ID and multimodal embeddings.

Dataset	#Users	#Items	#Interactions	Sparsity
Baby	19,445	7,050	160,792	99.8827
Sports	35,598	18,357	296,337	99.9547
Office	4,905	2,420	53,258	99.5513
Video Games	24,303	10,672	231,780	99.9106

Table 1. Dataset statistics

Dataset	Metric	Accuracy-oriented MMRSs					Alignment-based MMRSs
		VBPR	LATTICE	FREEDOM	LGMRec	SMORE	BM3	DA-MRS	AlignRec	AnchorRec
Baby	R@20	0.0600	0.0859	0.0976	0.0980	0.0997	0.0831	0.0947	0.1007	0.1013
Baby	N@20	0.0261	0.0375	0.0420	0.0436	0.0443	0.0357	0.0420	0.0444	0.0436
Sports	R@20	0.0730	0.0950	0.1034	0.1068	0.1135	0.0907	0.1073	0.1135	0.1048
Sports	N@20	0.0329	0.0424	0.0455	0.0479	0.0508	0.0400	0.0480	0.0507	0.0458
Office	R@20	0.1069	0.1385	0.1466	0.1474	0.1619	0.1144	0.1475	0.1446	0.1513

Table 2. Overall top-

\boldsymbol{N}

recommendation performance across four Amazon datasets

Figure 4. Neighborhood overlap across embedding spaces.

Figure 5. Top-3 neighbors retrieved by AlignRec and \ours for the same target item: (a) quantitative comparison using 2-hop proximity and text and vision similarity, and (b) qualitative comparison using images and textual keywords.

Multimodal Recommendation

Multimodal recommendation combines user behavior with item content such as images, text, audio, and video to personalize rankings.

See all 48 papers

EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation
Aug 22, 2025Xiaoxiong Zhang, Xin Zhou, Zhiwei Zeng +2Cross-Modal Alignment Multimodal Embedding
College of Computing and Data Science, Nanyang Technological University, Singapore
MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning
Sep 4, 2026Ahmad Mousavi, Majid Alikhani, Yeon-Chang Lee +2Multimodal Recommendation Multimodal Graph Learning
Department of Mathematics and Statistics American University · Independent Researcher Toronto, Ontario, Canada · Independent Researcher +4
Sequential Modality Dropout for Robust Multi-Modal Sequential Recommendation
Aug 10, 2026Guanqun Yang, Wenlong ZhangMissing-Modality Learning Multimodal Robustness
Stevens Institute of Technology Hoboken, NJ, USA

Aug 22, 2025J/K move · Enter open · S save

EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation

Xiaoxiong Zhang, Xin Zhou, Zhiwei Zeng, Yongjie Wang+1

College of Computing and Data Science, Nanyang Technological University, Singapore

Cross-Modal Alignment Multimodal Embedding Recommender Systems Multimodal Recommendation Multimodal Graph Learning

Open paper AskPDF arXiv

MultiModal Recommendation (MMR) systems have emerged as a promising solution for improving recommendation quality by leveraging rich item-side modality information, prompting a surge of diverse methods. Despite these advances, existing methods still face two critical limitations. First, they use raw modality features to construct item-item links for enriching the behavior graph, while giving limited attention to balancing collaborative and modality-aware semantics or mitigating modality noise in the process. Second, they use a uniform alignment weight across all entities and also maintain a fixed alignment strength throughout training, limiting the effectiveness of modality-behavior alignment. To address these challenges, we propose EGRA. First, instead of relying on raw modality features, it alleviates sparsity by incorporating into the behavior graph an item-item graph built from representations generated by a pretrained MMR model. This enables the graph to capture both collaborative patterns and modality aware similarities with enhanced robustness against modality noise. Moreover, it introduces a novel bi-level dynamic alignment weighting mechanism to improve modality-behavior representation alignment, which dynamically assigns alignment strength across entities according to their alignment degree, while gradually increasing the overall alignment intensity throughout training. Extensive experiments on five datasets show that EGRA significantly outperforms recent methods, confirming its effectiveness.

Abstract

Multimodal Recommendation

Explore similar work

EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation

MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning

Sequential Modality Dropout for Robust Multi-Modal Sequential Recommendation