Rethinking Contrastive Learning for Graph Collaborative Filtering: Limitations and a Simple Remedy

2026-05-19

AI Summary

This work addresses a key limitation in existing contrastive learning approaches for graph-based collaborative filtering—such as Sampled Softmax—which lack mechanisms to differentially weight neighbor pairs, thereby hindering effective user–item prediction. The study reveals that recommendation performance critically depends on selectively emphasizing neighbor pairs that are structurally similar to the target user and item. Building on this insight, the authors propose a novel contrastive learning objective, NT-SSM, which incorporates a type-aware weighting mechanism for neighbor pairs, overcoming the constraints of conventional dynamic updating strategies. Extensive experiments demonstrate that NT-SSM consistently outperforms the original SSM across multiple datasets and state-of-the-art graph collaborative filtering models, yielding significant improvements in recommendation accuracy.

Abstract

Graph collaborative filtering (GCF) is a dominant paradigm in recommender systems, where contrastive learning (CL) objectives such as the Sampled Softmax (SSM) loss are widely used for optimization. However, it remains unclear how CL interacts with the prediction mechanism of GCF. By unfolding the prediction mechanism of GCF, we show that the user-item prediction score is computed by aggregating learnable weights over a large number of neighbor pairs formed by the multi-hop neighbors of the user and the item. This analysis suggests that effective optimization critically depends on which neighbor pairs are upweighted during training. Empirically, we find that effective recommendation is achievable by selectively upweighting only a small subset of neighbor pairs whose constituent neighbors are structurally similar to the target user and item, and that the effect of such selective upweighting varies across different neighbor pair types. Based on these findings, we analyze SSM and identify key limitations in its neighbor pair weight update dynamics. To address these limitations, we propose NT-SSM, an effective and principled CL objective that induces type-aware neighbor pair weight update dynamics. Experiments demonstrate consistent performance improvements over SSM across multiple datasets and GCF models.

Problem

• Graph Collaborative Filtering
• Contrastive Learning
• Neighbor Pair Weighting
• Sampled Softmax
• Recommendation Systems

Innovation

• Graph Collaborative Filtering
• Contrastive Learning
• Neighbor Pair Weighting
• Type-Aware Optimization
• Sampled Softmax