Long-History User Transformers for Real-Time Ad Ranking

• URL: http://arxiv.org/abs/2607.14331v1
• Date: Wed, 15 Jul 2026 19:51:16 GMT
• Title: Long-History User Transformers for Real-Time Ad Ranking
• Abstract summary: We describe how a production advertising system resolves this conflict by decoupling history encoding from real-time inference.<n>A high-capacity offline transformer asynchronously encodes the user's full cross-surface interaction history into a compact representation cached in a feature store.<n>In production A/B experiments, the system improves the primary ranking metric by +2.77% in search advertising and +2.1% on the Yandex Advertising Network, with revenue gains of +2.26% and +0.43% respectively.
• Score: 0.027185251060695432
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long interaction histories are among the most informative inputs for click-through rate (CTR) prediction, yet in online advertising they collide with a hard serving constraint: ads must be scored within a few hundred milliseconds to enter the auction, which rules out running a large sequence encoder at request time. We describe how a production advertising system resolves this conflict by decoupling history encoding from real-time inference. A high-capacity offline transformer asynchronously encodes the user's full cross-surface interaction history into a compact representation cached in a feature store, while a lightweight runtime model combines this cached representation with the user's most recent events and the request context at serving time. The offline encoder is pre-trained autoregressively on large-scale interaction logs with a dual objective - feedback prediction and next-item prediction - and the two-stage architecture is then fine-tuned for CTR prediction on the target advertising surface. Offline, the split design recovers 72-80% of the quality of a full-history runtime transformer that would be too expensive to deploy, and the cached representation is robust enough to staleness to permit inexpensive refresh policies. In production A/B experiments, the system improves the primary ranking metric by +2.77% in search advertising and +2.1% on the Yandex Advertising Network, with revenue gains of +2.26% and +0.43% respectively - without increasing serving latency.

Related papers

• ChronicleRec: Pre-training Temporally Anchored Tokens for Lifelong User Modeling [19.519280627157716]
  Existing lifelong-interest methods retrieve target-relevant behaviors for each candidate.<n>Recent target-independent compression methods enable cached user summaries, but often append query tokens at the sequence end.<n>We propose ChronicleRec, a pre-train-and-transfer framework that compresses an ultra-long behavior sequence once into a chronologically ordered set of Chronicle Tokens.
  arXiv  Detail & Related papers  (2026-09-11T02:47:55Z)
• GRACE: Generative Recommender Acceleration Engine for Real-Time Ads Retrieval [15.73766250149643]
  This paper presents GRACE, a serving system for ads generative retrieval.<n>For eligibility, GRACE introduces Generative Target Matching (GTM), which extends catalog-valid constrained decoding.<n>For compute-cost and latency, GRACE targets encoder-decoder Transformers, which are more lightweight than LLMs.
  arXiv  Detail & Related papers  (2026-08-02T02:30:38Z)
• TransX: Scaling Transformer-based Recommendation via Behavioral and Serving Stream Crossings [4.750812376477589]
  We propose TransX, a production-oriented encoder-decoder architecture that reformulates recommendation as a sequence-to-sequence action problem.<n>To enable low-latency, high-QPS deployment, TransX is co-designed with an amortized serving strategy that combines incremental behavior encoding with per-request key-value caching.
  arXiv  Detail & Related papers  (2026-07-31T01:42:57Z)
• Building a User Foundation Model for the Open Web [2.097346573871652]
  We present a user foundation model that applies self-supervised learning on user browsing histories.<n>We show that the learned representation improves multiple downstream production tasks.
  arXiv  Detail & Related papers  (2026-07-30T11:07:54Z)
• Back from the Future: Key-Value Cache Management by Counter-Causal Surprise [67.1056509495879]
  Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years.<n>We propose a simple yet effective KV eviction scheme motivated by the insight that past tokens which can be well-predicted from more recent tokens are redundant.<n>We evaluate our strategy on various open-source LLMs and benchmark datasets showing competitive or improved performance over other state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-07-30T02:42:51Z)
• DYNA-PRUNER: Input-Adaptive Data-Model Co-Pruning for Efficient and Scalable Spatio-Temporal Media Prediction [24.16400471463046]
  Dyna-Pruner is an end-to-end framework for input-dependent co-pruning of data and model structure.<n>A shared-importance synchronization mechanism generates coupled masks that prune redundant regions.<n>Experiments on WeatherBench, SEVIR, and TaxiBJ show seamless integration with CNN, RNN, and Transformer backbones.
  arXiv  Detail & Related papers  (2026-06-13T15:14:05Z)
• End-to-End Context Compression at Scale [81.70601323130997]
  Long-context language model inference is bottlenecked by memory, as the KV cache grows with context length.<n>Recent techniques to compress the KV cache fall short: they either degrade model quality substantially or require considerable time and compute to compress a single long prompt.<n>Existing approaches are not competitive with KV cache compression on the accuracy-efficiency frontier.
  arXiv  Detail & Related papers  (2026-06-08T15:43:16Z)
• Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion [61.57938553036056]
  ARL2 is a hybrid attention module that replaces quadratic cross-frame attention with a fixed-size recurrent state.<n>We decompose self-attention into two branches: an intra-frame softmax branch for spatial detail and local dependencies, and an inter-frame recurrent linear branch that maintains a fixed-size state for streaming context.<n>With 75% of layers replaced by hybrid linear attention, the model achieves up to 2.26 wall-clock speedup and 54% memory reduction, while maintaining comparable quality with improving temporal consistency.
  arXiv  Detail & Related papers  (2026-05-15T19:33:45Z)
• Beyond Item IDs: Scaling Short-Form-Video Recommendation via Semantic-Native Long Sequence Modeling [13.58338102892064]
  We present a production-deployed framework for modeling ultra-long user behavior sequences at a billion-user scale.<n>We first address the representation bottleneck by adopting content-native Semantic IDs.<n>Second, to overcome the sequence scaling barrier, we introduce a Global-Aware Compression Transformer.
  arXiv  Detail & Related papers  (2026-05-04T00:00:53Z)
• CADET: Context-Conditioned Ads CTR Prediction With a Decoder-Only Transformer [11.143142950626169]
  We present CADET, an end-to-end decoder-only transformer for ads CTR prediction deployed at LinkedIn.<n>In online A/B testing, CADET achieves a 11.04% CTR lift compared to the production LiRank baseline model.<n>The system has been successfully deployed on LinkedIn's advertising platform, serving the main traffic for homefeed sponsored updates.
  arXiv  Detail & Related papers  (2026-02-11T22:24:33Z)
• VideoCompressa: Data-Efficient Video Understanding via Joint Temporal Compression and Spatial Reconstruction [55.66673587952058]
  Video understanding models are increasingly limited by the prohibitive storage and computational costs of large-scale datasets.<n>VideoCompressa is a novel framework for video data synthesis that reframes the problem as dynamic latent compression.
  arXiv  Detail & Related papers  (2025-11-24T07:07:58Z)
• Bidding-Aware Retrieval for Multi-Stage Consistency in Online Advertising [30.108437268612438]
  Bidding-Aware Retrieval (BAR) is a model-based retrieval framework that addresses multi-stage inconsistency by incorporating ad bid value into the retrieval scoring function.<n>BAR's core innovation is Bidding-Aware Modeling, incorporating bid signals through monotonicity-constrained learning and multi-task distillation to ensure economically coherent representations.<n>Extensive offline experiments and full-scale deployment across Alibaba's display advertising platform validated BAR's efficacy.
  arXiv  Detail & Related papers  (2025-08-07T09:43:34Z)
• FastCar: Cache Attentive Replay for Fast Auto-Regressive Video Generation on the Edge [60.000984252907195]
  Auto-regressive (AR) models have recently shown promise in visual generation tasks due to their superior sampling efficiency.<n>Video generation requires a substantially larger number of tokens to produce coherent temporal frames, resulting in significant overhead during the decoding phase.<n>We propose the textbfFastCar framework to accelerate the decode phase for the AR video generation by exploring the temporal redundancy.
  arXiv  Detail & Related papers  (2025-05-17T05:00:39Z)
• HiRE: High Recall Approximate Top-$k$ Estimation for Efficient LLM Inference [68.59839755875252]
  HiRE comprises of two novel components: (i) a compression scheme to cheaply predict top-$k$ rows/columns with high recall, followed by full computation restricted to the predicted subset, and (ii) DA-TOP-$k$: an efficient multi-device approximate top-$k$ operator. We demonstrate that on a one billion parameter model, HiRE applied to both the softmax as well as feedforward layers, achieves almost matching pretraining and downstream accuracy, and speeds up inference latency by $1.47times$ on a single TPUv5e device.
  arXiv  Detail & Related papers  (2024-02-14T18:04:36Z)
• ReBotNet: Fast Real-time Video Enhancement [59.08038313427057]
  Most restoration networks are slow, have high computational bottleneck, and can't be used for real-time video enhancement. In this work, we design an efficient and fast framework to perform real-time enhancement for practical use-cases like live video calls and video streams. To evaluate our method, we emulate two new datasets that real-world video call and streaming scenarios, and show extensive results on multiple datasets where ReBotNet outperforms existing approaches with lower computations, reduced memory requirements, and faster inference time.
  arXiv  Detail & Related papers  (2023-03-23T17:58:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.