Unified Semantic and ID Representation Learning for Deep Recommenders

2025-02-23

AI Summary

To address weak generalization and poor cold-start performance in recommender systems—caused by redundant ID representations and unstable semantic representations—this paper proposes a joint representation learning framework for ID and semantic tokens. We introduce a novel dual-channel token modeling mechanism, theoretically and empirically revealing the hierarchical advantages of cosine similarity for embedding disentanglement and Euclidean distance for discriminative decision-making, and accordingly design a hybrid distance metric architecture. Our method employs deep neural networks to jointly optimize ID tokens and semantic tokens in an end-to-end manner. Extensive experiments on three benchmark datasets demonstrate that our approach outperforms state-of-the-art models by 6%–17% in recommendation accuracy, reduces token parameter count by over 80%, and significantly improves cold-start and long-tail item recommendation performance.

Technology Category

• Machine Learning: Representation Learning
• Data Mining & Knowledge Management: Recommender Systems
• Computer Vision: Representation Learning for Vision

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Effective recommendation is crucial for large-scale online platforms. Traditional recommendation systems primarily rely on ID tokens to uniquely identify items, which can effectively capture specific item relationships but suffer from issues such as redundancy and poor performance in cold-start scenarios. Recent approaches have explored using semantic tokens as an alternative, yet they face challenges, including item duplication and inconsistent performance gains, leaving the potential advantages of semantic tokens inadequately examined. To address these limitations, we propose a Unified Semantic and ID Representation Learning framework that leverages the complementary strengths of both token types. In our framework, ID tokens capture unique item attributes, while semantic tokens represent shared, transferable characteristics. Additionally, we analyze the role of cosine similarity and Euclidean distance in embedding search, revealing that cosine similarity is more effective in decoupling accumulated embeddings, while Euclidean distance excels in distinguishing unique items. Our framework integrates cosine similarity in earlier layers and Euclidean distance in the final layer to optimize representation learning. Experiments on three benchmark datasets show that our method significantly outperforms state-of-the-art baselines, with improvements ranging from 6% to 17% and a reduction in token size by over 80%. These results demonstrate the effectiveness of combining ID and semantic tokenization to enhance the generalization ability of recommender systems.

Problem

• Combining ID and semantic tokens for recommendations
• Optimizing embedding search with cosine and Euclidean metrics
• Enhancing recommender system generalization and efficiency

Innovation

• Unified Semantic and ID Representation Learning
• Cosine similarity decouples embeddings effectively
• Combines ID and semantic tokens for better recommendations