IntSR: An Integrated Generative Framework for Search and Recommendation

2025-09-25

AI Summary

Existing generative recommender systems primarily unify retrieval and ranking while neglecting the organic integration of search (explicit queries) and recommendation (implicit interests), and struggle with erroneous pattern learning and escalating computational complexity caused by dynamic item catalogs. This paper proposes IntSR—the first generative framework unifying search and recommendation—centered on user queries to jointly model multimodal explicit requests and implicit interest representations, integrated with an efficient retrieval mechanism for joint optimization of search, recommendation, and ranking. Evaluated in multiple real-world scenarios on AutoNavi Map, IntSR achieves a 3.02% increase in digital asset GMV, a 2.76% uplift in POI recommendation CTR, and a 5.13% improvement in travel mode suggestion accuracy.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Generative recommendation has emerged as a promising paradigm, demonstrating remarkable results in both academic benchmarks and industrial applications. However, existing systems predominantly focus on unifying retrieval and ranking while neglecting the integration of search and recommendation (S&R) tasks. What makes search and recommendation different is how queries are formed: search uses explicit user requests, while recommendation relies on implicit user interests. As for retrieval versus ranking, the distinction comes down to whether the queries are the target items themselves. Recognizing the query as central element, we propose IntSR, an integrated generative framework for S&R. IntSR integrates these disparate tasks using distinct query modalities. It also addresses the increased computational complexity associated with integrated S&R behaviors and the erroneous pattern learning introduced by a dynamically changing corpus. IntSR has been successfully deployed across various scenarios in Amap, leading to substantial improvements in digital asset's GMV(+3.02%), POI recommendation's CTR(+2.76%), and travel mode suggestion's ACC(+5.13%).

Problem

• Integrating search and recommendation tasks using distinct query modalities
• Addressing computational complexity from integrated search-recommendation behaviors
• Solving erroneous pattern learning from dynamically changing corpus

Innovation

• Integrated generative framework using distinct query modalities
• Addresses computational complexity of integrated search-recommendation behaviors
• Mitigates erroneous pattern learning from dynamically changing corpus