Search papers, labs, and topics across Lattice.
This paper introduces UniCon, a unified context-centric modeling architecture for click-through rate (CTR) prediction that addresses the limitations of existing approaches by treating both historical behavior and current requests as homogeneous context units. By employing intra-context and inter-context attention mechanisms, UniCon effectively captures local item relationships and the dynamic evolution of decision states, leading to improved scaling efficiency and prediction quality. The model demonstrates significant performance gains, achieving a 0.0139 increase in offline AUC and notable online improvements in revenue and CTR metrics on Meituan search advertising.
Treating historical behavior and current requests as homogeneous context units, UniCon achieves a 3.09% lift in revenue and 2.07% in CTR, outperforming traditional models.
Unified modeling has become a major direction for industrial click-through rate (CTR) prediction. Existing approaches typically unify sequential and non-sequential signals at the token level, model their interactions in a shared backbone, and increase model capacity to improve scaling behavior. However, this division originates from legacy feature-engineering practice and is misaligned with the underlying decision process. User behavior is inherently a sequence of homogeneous context units; at the level of input organization, historical behavior and the current request differ only in whether their outcomes are observed or remain to be predicted. Treating them as heterogeneous signals obscures structural dependencies within the user's decision context, limiting both scaling efficiency and prediction quality. This limitation is particularly pronounced in context-rich scenarios such as e-commerce shelves and waterfall feeds. To address this, we propose UniCon, a unified context-centric modeling architecture that treats the request context as the basic modeling unit and organizes history and prediction targets as homogeneous context units. Intra-context attention captures local coupling among items within a context (Locality), while inter-context attention models the dynamic evolution of decision states across contexts (Dynamics). This organization bridges the structural gap between history and target and supports more effective scaling of unified CTR models. Context-unit-level sequence compression further reduces deployment overhead. On Meituan search advertising, UniCon improves offline AUC by 0.0139 over a strong production baseline and achieves statistically significant online lifts of 3.09% in RPM, 2.07% in CTR, and 2.95% in revenue.
