LENS: A Staged Design for Interaction Granularityin Sequential CTR Prediction
2026-05-25 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors explore how to best link a target item with a user's behavior sequence for predicting click-through rates (CTR). Existing methods either score each item directly, which struggles with rare items, or use combined queries that lose some target-specific detail. They introduce LENS, a system with two modules that bring back target-focused control in coarser query setups, plus a position-aware tool called QueryPos. Tested on several models and datasets, their approach consistently improved performance. They also found that when items are rare, conditioning on both the item and the sequence works better than using just the item.
CTR predictionsequential recommendationitem embeddingslatent-query architecturestarget-conditioned queryposition biasuser behavior sequencedata sparsityquery representation
Authors
Yuan Wang, Yue Liu, Jun Zhang, Jie Jiang
Abstract
In sequential CTR prediction, a central design question is at what granularity the target should interact with the user behaviour sequence. Existing models mainly follow two routes. Raw-item architectures such as DIN let the target score each item in the sequence directly. This relies on well-trained item embeddings and becomes brittle for sparse items. Latent-query architectures such as HyFormer, MixFormer, and OneTrans build query representations by combining the target with other information. This is more robust across item-density regimes but blunter: target-specific control is diluted. We propose LENS to restore target-specific control within these coarser bottlenecks. LENS has two modules: a Target-Conditioned Query Gate (TCQG) for query activation and a Target-Conditioned Position Bias (TCPB) for history retrieval. We further introduce Query-Specific Position Bias (QueryPos), a simple static position-aware reference for latent-query backbones. Across three representative latent-query backbones and four datasets, the combined QueryPos+LENS design achieves positive total-gain point estimates in all twelve evaluated backbone--dataset cells. We also identify a density-dependent conditioning rule: as item density decreases, the optimal condition source shifts from item-only to item-plus-sequence.