SIGMA: A Semantic-Grounded Instruction-Driven Generative Multi-Task Recommender at AliExpress

2026-02-26

AI Summary

This work proposes SIGMA, a novel generative recommendation framework that overcomes the limitations of existing interaction-driven next-item prediction systems, which struggle to adapt to dynamic trends and diverse task requirements. SIGMA constructs a semantically unified latent space to integrate collaborative and semantic information, employs a hybrid item tokenization strategy, and leverages a large-scale multi-task instruction-tuning dataset to enable instruction-driven generative recommendations. The model implements a three-stage generation pipeline combined with an adaptive probability fusion mechanism. Extensive offline evaluations and online A/B tests demonstrate that SIGMA significantly improves both recommendation accuracy and diversity, offering a flexible and efficient solution for multi-task recommendation scenarios.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

With the rapid evolution of Large Language Models, generative recommendation is gradually reshaping the paradigm of recommender systems. However, most existing methods are still confined to the interaction-driven next-item prediction paradigm, failing to rapidly adapt to evolving trends or address diverse recommendation tasks along with business-specific requirements in real-world scenarios. To this end, we present SIGMA, a Semantic-Grounded Instruction-Driven Generative Multi-Task Recommender at AliExpress. Specifically, we first ground item entities in general semantics via a unified latent space capturing both semantic and collaborative relations. Building upon this, we develop a hybrid item tokenization method for precise modeling and efficient generation. Moreover, we construct a large-scale multi-task SFT dataset to empower SIGMA to fulfill various recommendation demands via instruction-following. Finally, we design a three-step item generation procedure integrated with an adaptive probabilistic fusion mechanism to calibrate the output distributions based on task-specific requirements for recommendation accuracy and diversity. Extensive offline experiments and online A/B tests demonstrate the effectiveness of SIGMA.

Problem

• generative recommendation
• multi-task recommendation
• instruction-driven
• semantic grounding
• real-world scenarios

Innovation

• generative recommendation
• semantic grounding
• instruction-driven
• multi-task learning
• hybrid tokenization