aiqbench analysis
The AIR framework addresses the latency bottleneck of LLMs in real-time recommendation systems by shifting inference to an offline phase, achieving a 400x speedup while preserving semantic reasoning. Industrial validation from Kuaishou E-commerce confirms the approach delivers significant business value, including a +3.446% increase in GMV. This work demonstrates a viable path for deploying high-semantic models in latency-sensitive commercial environments.
Atomic Intent Reasoning: Bringing LLM Semantics to Industrial Cross-Domain Recommendations
arXiv:2606.10357v1 Announce Type: cross Abstract: Cross-domain recommendation is a core problem in content-to-e-commerce platforms. Its objective is to leverage user interactions with content to infer potential purchasing intent on the e-commerce side, thereby enhancing conversion rates and commercial value. However, in real industrial scenarios, cross-domain recommendation faces multiple challenges: significant semantic gaps exist between different domains, and user cross-domain behavior sequences are often massive in scale and rich in noise. Although large language models (LLMs) possess powerful semantic understanding and reasoning capabilities, their millisecond-level inference latency makes direct application in online recommendation systems difficult. To address these issues, this paper introduces AIR (Atomic Intent Reasoning), an LLM-driven cross-domain recommendation framework designed for industrial-grade deployment. By migrating LLM inference to the offline phase and dynamically constructing user intent representations through efficient retrieval and composition during online operations, it achieves approximately 400* inference acceleration while maintaining semantic consistency. Experimental results across multiple public datasets demonstrate that our method achieves state-of-the-art performance in cross-domain recommendation tasks. Furthermore, large-scale online A/B testing conducted in Kuaishou E-commerce's real-world business scenarios shows that our approach delivers stable and significant improvements across multiple core business metrics, including a +3.446% increase in GMV, fully validating its effectiveness and practical value in industrial-scale recommendation systems.
Abstract:Cross-domain recommendation is a core problem in content-to-e-commerce platforms. Its objective is to leverage user interactions with content to infer potential purchasing intent on the e-commerce side, thereby enhancing conversion rates and commercial value. However, in real industrial scenarios, cross-domain recommendation faces multiple challenges: significant semantic gaps exist between different domains, and user cross-domain behavior sequences are often massive in scale and rich in noise. Although large language models (LLMs) possess powerful semantic understanding and reasoning capabilities, their millisecond-level inference latency makes direct application in online recommendation systems difficult. To address these issues, this paper introduces AIR (Atomic Intent Reasoning), an LLM-driven cross-domain recommendation framework designed for industrial-grade deployment. By migrating LLM inference to the offline phase and dynamically constructing user intent representations through efficient retrieval and composition during online operations, it achieves approximately 400* inference acceleration while maintaining semantic consistency. Experimental results across multiple public datasets demonstrate that our method achieves state-of-the-art performance in cross-domain recommendation tasks. Furthermore, large-scale online A/B testing conducted in Kuaishou E-commerce's real-world business scenarios shows that our approach delivers stable and significant improvements across multiple core business metrics, including a +3.446% increase in GMV, fully validating its effectiveness and practical value in industrial-scale recommendation systems.
Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
arXiv:2606.10357 [cs.IR]
(or arXiv:2606.10357v1 [cs.IR] for this version)
https://doi.org/10.48550/arXiv.2606.10357 arXiv-issued DOI via DataCite (pending registration)
Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 (KDD '26), August 09--13, 2026, Jeju Island, Republic of Korea
https://doi.org/10.1145/3770855.3818320 DOI(s) linking to related resources
Full text under the publisher's open licence · 252 words