Search papers, labs, and topics across Lattice.
The paper introduces OneRanker, a unified generative model for industrial advertising recommendation that addresses the misalignment between interest objectives and business value, target-agnostic generation, and disconnection between generation and ranking. OneRanker employs a value-aware multi-task decoupling architecture, a coarse-to-fine collaborative target awareness mechanism using Fake Item Tokens, and input-output dual-side consistency guarantees with a Distribution Consistency (DC) Constraint Loss. Deployed on Tencent's WeiXin channels, OneRanker demonstrates a significant improvement in GMV (+1.34%), showcasing its industrial feasibility.
OneRanker tackles the optimization tension in generative recommendation systems by deeply integrating generation and ranking into a single architecture, achieving state-of-the-art performance in industrial advertising.
The end-to-end generative paradigm is revolutionizing advertising recommendation systems, driving a shift from traditional cascaded architectures towards unified modeling. However, practical deployment faces three core challenges: the misalignment between interest objectives and business value, the target-agnostic limitation of generative processes, and the disconnection between generation and ranking stages. Existing solutions often fall into a dilemma where single-stage fusion induces optimization tension, while stage decoupling causes irreversible information loss. To address this, we propose OneRanker, achieving architectural-level deep integration of generation and ranking. First, we design a value-aware multi-task decoupling architecture. By leveraging task token sequences and causal mask, we separate interest coverage and value optimization spaces within shared representations, effectively alleviating target conflicts. Second, we construct a coarse-to-fine collaborative target awareness mechanism, utilizing Fake Item Tokens for implicit awareness during generation and a ranking decoder for explicit value alignment at the candidate level. Finally, we propose input-output dual-side consistency guarantees. Through Key/Value pass-through mechanisms and Distribution Consistency (DC) Constraint Loss, we achieve end-to-end collaborative optimization between generation and ranking. The full deployment on Tencent's WeiXin channels advertising system has shown a significant improvement in key business metrics (GMV - Normal +1.34\%), providing a new paradigm with industrial feasibility for generative advertising recommendations.
