mlsanityData to 14 Oct 2026

Distribution-Aware End-to-End Embedding for Streaming Numerical Features in Click-Through Rate Prediction

arXiv 2602.03223 ↗First posted 3 Feb 2026cs.IR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper explores effective numerical feature embedding for Click-Through Rate prediction in streaming environments. Conventional static binning methods rely on offline statistics of numerical distributions; however, this inherently two-stage process often triggers semantic drift during bin boundary updates. While neural embedding methods enable end-to-end learning, they often discard explicit distributional information. Integrating such information end-to-end is challenging because streaming features often violate the i.i.d. assumption, precluding unbiased estimation of the population distribution via the expectation of order statistics. Furthermore, the critical context dependency of numerical distributions is often neglected. To this end, we propose DAES, an end-to-end framework designed to tackle numerical feature embedding in streaming training scenarios by integrating distributional information with an adaptive modulation mechanism. Specifically, we introduce an efficient reservoir-sampling-based distribution estimation method and two field-aware distribution modulation strategies to capture streaming distributions and field-dependent semantics. DAES significantly outperforms existing approaches as demonstrated by extensive offline and online experiments and has been fully deployed on a leading short-video platform with hundreds of millions of daily active users.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 42 of its 71 references that are in the map. First posted on arXiv 3 Feb 2026.

Builds on

42 of 71 references.

References · 42

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2016Wide & Deep Learning for Recommender Systemscited by 747
  3. 2021Revisiting Deep Learning Models for Tabular Datacited by 740
  4. 2017Deep & Cross Network for Ad Click Predictionscited by 393
  5. 2018Deep Interest Evolution Network for Click-Through Rate Predictioncited by 338
  6. 2019Deep Learning Recommendation Model for Personalization and Recommendation Systemscited by 327
  7. 2020DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systemscited by 327
  8. 2018AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networkscited by 320
  9. 2018xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systemscited by 278
  10. 2022On Embeddings for Numerical Features in Tabular Deep Learningcited by 186
  11. 2017Neural Factorization Machines for Sparse Predictive Analyticscited by 183
  12. 2017Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networkscited by 126
30 more
  1. 2016Deep Learning over Multi-field Categorical Data: A Case Study on User Response Predictioncited by 67
  2. 2018Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertisingcited by 59
  3. 2019Adaptive Factorization Network: Learning Adaptive-Order Feature Interactionscited by 48
  4. 2021MaskNet: Introducing Feature-Wise Multiplication to CTR Ranking Models by Instance-Guided Maskcited by 48
  5. 2016Higher-Order Factorization Machinescited by 45
  6. 2018Product-based Neural Networks for User Response Prediction over Multi-field Categorical Datacited by 44
  7. 2019Neural Input Search for Large Scale Recommendation Modelscited by 39
  8. 2020An Embedding Learning Framework for Numerical Features in CTR Predictioncited by 37
  9. 2022Towards Understanding the Overfitting Phenomenon of Deep Click-Through Rate Prediction Modelscited by 33
  10. 2023Personalized Graph Signal Processing for Collaborative Filteringcited by 29
  11. 2025AgentCF++: Memory-enhanced LLM-based Agents for Popularity-aware Cross-domain Recommendationscited by 24
  12. 2020Network On Network for Tabular Data Classification in Real-world Applicationscited by 17
  13. 2022Enhancing CTR Prediction with Context-Aware Feature Representation Learningcited by 17
  14. 2023AutoSeqRec: Autoencoder for Efficient Sequential Recommendationcited by 15
  15. 2024Ads Recommendation in a Collapsed and Entangled Worldcited by 13
  16. 2022Parameter-free Dynamic Graph Embedding for Link Predictioncited by 12
  17. 2023Triple Structural Information Modelling for Accurate, Explainable and Interactive Recommendationcited by 12
  18. 2019Computing Extremely Accurate Quantiles Using t-Digestscited by 10
  19. 2019Operation-aware Neural Networks for User Response Predictioncited by 10
  20. 2025Improving LLM-powered Recommendations with Personalized Informationcited by 10
  21. 2021XCrossNet: Feature Structure-Oriented Learning for Click-Through Rate Predictioncited by 6
  22. 2025Unbiased Collaborative Filtering with Fair Samplingcited by 6
  23. 2021DCAP: Deep Cross Attentional Product Network for User Response Predictioncited by 5
  24. 2023Function Basis Encoding of Numerical Features in Factorization Machinescited by 4
  25. 2024Filtering Discomforting Recommendations with Large Language Modelscited by 4
  26. 2025FedCIA: Federated Collaborative Information Aggregation for Privacy-Preserving Recommendationcited by 4
  27. 2025Bidirectional Knowledge Distillation for Enhancing Sequential Recommendation with Large Language Modelscited by 4
  28. 2026Dynamic Feature-Embedding Communication via Codebook Distillation for Federated Recommendationcited by 3
  29. 2025Enhancing Interpretability and Effectiveness in Recommendation with Numerical Features via Learning to Contrast the Counterfactual samplescited by 2
  30. 2021Online Product Feature Recommendations with Interpretable Machine Learningcited by 1

Built on by

Not cited yet.