发表机构
Z Lab（Z实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对KDD Cup 2026腾讯UniRec挑战赛，我们通过稠密特征表示与优化提升CVR预测AUC至0.828535，并揭示序列建模贡献微小及验证集高估风险。
AI 中文摘要
我们描述了在KDD Cup 2026腾讯UniRec挑战赛中获得第10名的解决方案，该任务是在3482万条记录上进行工业级点击到转化（CVR）预测，并探究哪些机制真正提升了留出集上的AUC。从官方PCVRHyFormer基线出发，通过15步单变量修改链将测试AUC从0.813237提升至0.827816，最终提交达到0.828535。对完整模型进行留一消融实验以归因提升：移除稠密特征表示栈损失0.0095 AUC，移除正交化优化器损失0.0028，而任何序列建模组件（合并的单流主干、极性通道、辅助头、逐token FFN）的损失均不超过0.0005，处于或接近±0.0004的随机种子波动带。我们还报告了一个泛化风险：按行分组的训练/验证划分共享同一时间窗口，因此验证AUC比排行榜高估约0.014；针对反记忆和高基数ID的修改甚至会在验证集上出现符号反转，这种分歧源于转储间的分布偏移，并且在按时间排序的重新划分后依然存在。在此规模下，稠密表示和优化而非更精细的序列建模驱动CVR AUC，最终结论必须依据留出排行榜。
英文摘要
We describe our 10th-place solution to the KDD Cup 2026 Tencent UniRec Challenge, industrial click-to-conversion (CVR) prediction over 34.82M records, and we ask which mechanisms actually move held-out AUC. Starting from the official PCVRHyFormer baseline, a 15-step single-variable chain raises test AUC from 0.813237 to 0.827816, and our final submission reaches 0.828535. A leave-one-out ablation from the full model attributes the gain: removing the dense-feature representation stack costs 0.0095 AUC and removing the orthogonalized optimizer costs 0.0028, while no sequence-modeling component (merged single-stream backbone, polarity channel, auxiliary head, per-token FFN) costs more than 0.0005, within or adjacent to a $\pm$0.0004 seed band. We also report a generalization hazard: the row-group train/validation split shares one time window, so validation AUC overstates the leaderboard by about 0.014; anti-memorization and high-cardinality-ID changes even invert sign against it, a divergence that traces to dump-to-dump distribution shift and survives a time-ordered re-split. Dense representation and optimization, not finer sequence modeling, drive CVR AUC at this scale, and verdicts must come from the held-out leaderboard.
Comments6 pages, 1 figure, 4 tables. KDD Cup 2026 Tencent UniRec Challenge Workshop
