抖音的生成式端到端广告检索
Generative End-to-end Ad Retrieval at Douyin
- ByteDance（字节跳动）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对生成式检索中的表示坍缩与物品冲突耦合问题，提出GEAR框架，联合优化分词器、生成器与重排序器，通过BasisVQ/BasisRQ稳定训练并消除冲突，已在抖音广告大规模落地。
AI中文摘要：
生成式检索将推荐重新定义为离散物品令牌的生成。然而，将这一范式扩展到现实世界的推荐系统时，暴露出两个关键瓶颈：1) 表示坍缩，即物品分词器在连续分布偏移下收敛到退化结果，从根本上阻碍了稳定的端到端适应。2) 物品冲突，即庞大的候选池导致不同物品共享相同的令牌序列，损害了最终检索精度。至关重要的是，这些瓶颈本质上是耦合的：扩大码本容量以缓解冲突不可避免地会加剧坍缩。为了同时解决这些问题，我们提出了GEAR，一个联合优化分词器、生成器和重排序器的端到端框架。为了缓解表示坍缩，我们引入了BasisVQ，它通过正交基重新参数化码本，实现全局梯度共享和潜在空间的刚性空间旋转，有效稳定梯度动态，无需临时启发式方法。我们进一步将其扩展为前缀感知的BasisRQ，在相同渐近时间复杂度下显著增强码本的表达能力。为了解决物品冲突，GEAR在生成过程中集成了一个上下文条件重排序头，以最小计算开销高效消除冲突物品的歧义。通过在一个端到端生成框架内统一稳定的分词和联合重排序，GEAR建立了一个完全可微分且可扩展的范式。它目前服务于抖音广告上的数亿日活跃用户，在广泛的在线A/B测试中产生了显著的实证改进。
英文摘要：
Generative retrieval reformulates recommendation as the generation of discrete item tokens. However, scaling this paradigm to real-world recommender systems reveals two critical bottlenecks: 1) Representation collapse, where the item tokenizer converges to degenerate results under continuous distribution shifts, fundamentally hindering stable end-to-end adaptation. 2) Item collisions, where the massive candidate pool causes distinct items to share identical token sequences, compromising the final retrieval precision. Crucially, these bottlenecks are inherently coupled: expanding codebook capacity to mitigate collisions inevitably exacerbates collapse. To address them simultaneously, we propose GEAR, an end-to-end framework that jointly optimizes the tokenizer, generator, and reranker. To mitigate representation collapse, we introduce BasisVQ, which re-parameterizes the codebook via an orthogonal basis to enable global gradient sharing and rigid spatial rotation of the latent space, effectively stabilizing gradient dynamics without ad-hoc heuristics. We further extend it to prefix-aware BasisRQ, substantially enhancing the codebook's expressiveness with the same asymptotic time complexity. To resolve item collisions, GEAR integrates a context-conditioned reranking head into the generative process, efficiently disambiguating colliding items with minimal computational overhead. By unifying stable tokenization and joint reranking within an end-to-end generative framework, GEAR establishes a fully differentiable and scalable paradigm. It currently serves hundreds of millions of daily active users on Douyin Ads, yielding substantial empirical improvements in extensive online A/B tests.
