GateSID: Adaptive Gating for Semantic-Collaborative Alignment in Cold-Start Recommendation

2026-03-24

AI Summary

This work addresses the Matthew effect and declining platform diversity in recommender systems caused by the scarcity of collaborative signals for cold-start items. To mitigate this, we propose an item maturity–aware adaptive gating mechanism that dynamically fuses semantic and collaborative signals: emphasizing semantic-behavior consistency during the cold-start phase and prioritizing reliable collaborative signals for mature items. Our approach integrates a residual quantized VAE to generate hierarchical semantic IDs and introduces a gating-controlled contrastive alignment module coupled with a shared attention mechanism. Evaluated on large-scale industrial datasets, the method significantly outperforms strong baselines. Online A/B tests demonstrate a 2.6% increase in GMV, a 1.1% improvement in CTR, and a 1.6% rise in order volume, with less than 5ms added inference latency.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

In cold-start scenarios, the scarcity of collaborative signals for new items exacerbates the Matthew effect, which undermines platform diversity and remains a persistent challenge in real-world recommender systems. Existing methods typically enhance collaborative signals with semantic information, but they often suffer from a collaborative-semantic tradeoff: collaborative signals are effective for popular items but unreliable for cold-start items, whereas over-reliance on semantic information may obscure meaningful collaborative differences. To address this issue, we propose GateSID, a framework that uses an adaptive gating network to dynamically balance semantic and collaborative signals according to item maturity. Specifically, we first discretize multimodal features into hierarchical Semantic IDs using Residual Quantized VAE. Building on this representation, we design two key components: (1) Gating-Fused Shared Attention, which fuses intra-modal attention distributions with item-level gating weights derived from embeddings and statistical features; and (2) Gate-Regulated Contrastive Alignment, which adaptively calibrates cross-modal alignment, enforcing stronger semantic-behavior consistency for cold-start items while relaxing the constraint for popular items to preserve reliable collaborative signals. Extensive offline experiments on large-scale industrial datasets demonstrate that GateSID consistently outperforms strong baselines. Online A/B tests further confirm its practical value, yielding +2.6% GMV, +1.1% CTR, and +1.6% orders with less than 5 ms additional latency.

Problem

• cold-start recommendation
• collaborative-semantic tradeoff
• Matthew effect
• item diversity
• collaborative signals

Innovation

• adaptive gating
• semantic-collaborative alignment
• cold-start recommendation
• Residual Quantized VAE
• contrastive alignment