cs.LGFeb 11, 2026

CADET: Context-Conditioned Ads CTR Prediction With a Decoder-Only Transformer

Authors: David Pardoe, Neil Daftary, Miro Furtado, Aditya Aiyer, Yu Wang, Liuqing Li, Tao Song, Lars Hertel, 

Organizations: LinkedIn, Mountain View, California, USA

Abstract

Click-through rate (CTR) prediction is fundamental to online advertising systems. While Deep Learning Recommendation Models (DLRMs) with explicit feature interactions have long dominated this domain, recent advances in generative recommenders have shown promising results in content recommendation. However, adapting these transformer-based architectures to ads CTR prediction still presents unique challenges, including handling post-scoring contextual signals, maintaining offline-online consistency, and scaling to industrial workloads. We present CADET (Context-Conditioned Ads Decoder-Only Transformer), an end-to-end decoder-only transformer for ads CTR prediction deployed at LinkedIn. Our approach introduces several key innovations: (1) a context-conditioned decoding architecture with multi-tower prediction heads that explicitly model post-scoring signals such as ad position, resolving the chicken-and-egg problem between predicted CTR and ranking; (2) a self-gated attention mechanism that stabilizes training by adaptively regulating information flow at both representation and interaction levels; (3) a timestamp-based variant of Rotary Position Embedding (RoPE) that captures temporal relationships across timescales from seconds to months; (4) session masking strategies that prevent the model from learning dependencies on unavailable in-session events, addressing train-serve skew; and (5) production engineering techniques including tensor packing, sequence chunking, and custom Flash Attention kernels that enable efficient training and serving at scale. In online A/B testing, CADET achieves a 11.04% CTR lift compared to the production LiRank baseline model, a hybrid ensemble of DCNv2 and sequential encoders. The system has been successfully deployed on LinkedIn's advertising platform, serving the main traffic for homefeed sponsored updates.

Figures & tables

Explore similar work

CardsList
  1. DUET -- Dual User Embedding Transformers for Offsite Conversion Prediction

    Jun 8, 2026Reazul Hasan Russel, Mingwei Tang, Rostam Shirani +18Learning to RankUser Modeling

  2. LFM4Ads: Large Foundation Model for Ads Recommendation

    Aug 20, 2025Jie Jiang, Shangyu Zhang, Shijie Quan +32Recommender SystemsCross-Domain Recommendation

  3. Fine-Tuned LLM as a Complementary Predictor Improving Ads System

    May 27, 2026Hui Yang, Daiwei He, Kevin Jiang +20LLM Fine-TuningOnline Advertising