REVIEW 22 cited by
End-to-End User Behavior Retrieval in Click-Through RatePrediction Model
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
End-to-End User Behavior Retrieval in Click-Through RatePrediction Model
read the original abstract
Click-Through Rate (CTR) prediction is one of the core tasks in recommender systems (RS). It predicts a personalized click probability for each user-item pair. Recently, researchers have found that the performance of CTR model can be improved greatly by taking user behavior sequence into consideration, especially long-term user behavior sequence. The report on an e-commerce website shows that 23\% of users have more than 1000 clicks during the past 5 months. Though there are numerous works focus on modeling sequential user behaviors, few works can handle long-term user behavior sequence due to the strict inference time constraint in real world system. Two-stage methods are proposed to push the limit for better performance. At the first stage, an auxiliary task is designed to retrieve the top-$k$ similar items from long-term user behavior sequence. At the second stage, the classical attention mechanism is conducted between the candidate item and $k$ items selected in the first stage. However, information gap happens between retrieval stage and the main CTR task. This goal divergence can greatly diminishing the performance gain of long-term user sequence. In this paper, inspired by Reformer, we propose a locality-sensitive hashing (LSH) method called ETA (End-to-end Target Attention) which can greatly reduce the training and inference cost and make the end-to-end training with long-term user behavior sequence possible. Both offline and online experiments confirm the effectiveness of our model. We deploy ETA into a large-scale real world E-commerce system and achieve extra 3.1\% improvements on GMV (Gross Merchandise Value) compared to a two-stage long user sequence CTR model.
Forward citations
Cited by 22 Pith papers
-
Similar Users-Augmented Interest Network
SUIN improves CTR prediction by augmenting target user sequences with similar users' behaviors via embedding-based retrieval, user-specific position encoding, and user-aware target attention.
-
Sample Is Feature: Beyond Item-Level, Toward Sample-Level Tokens for Unified Large Recommender Models
SIF encodes full historical raw samples as tokens via hierarchical quantization to preserve sample context and unify sequential/non-sequential features in large recommender models.
-
Next-Scale Generative Reranking: A Tree-based Generative Rerank Method at Meituan
NSGR is a tree-structured generative reranker that progressively generates optimal lists via next-scale expansion and multi-scale neighbor loss to balance perspectives and align training signals.
-
SITA: Semantic Interest Tokens for Target-Aware Compression in Long-Sequence Recommendation
SITA learns semantic interest tokens per user and selects them with per-item semantic codes, giving target-aware long-sequence modeling at O(N) inference cost.
-
LENS: A Staged Design for Interaction Granularity in Sequential CTR Prediction
LENS restores target-specific control in latent-query CTR models via TCQG and TCPB modules plus QueryPos reference, reporting positive gains in all 12 backbone-dataset cells and a density-dependent conditioning rule.
-
Generative Long-term User Interest Modeling for Click-Through Rate Prediction
GenLI generates diverse target-independent interest distributions via an IGM, retrieves behaviors with O(1) lookup in BRM, and fuses via IFM gating to balance accuracy and efficiency in CTR prediction.
-
UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
UxSID uses Semantic IDs and dual-level attention for semantic-group shared interest memory to efficiently model ultra-long user sequences, claiming SOTA performance and 0.337% revenue lift in advertising A/B tests.
-
Versioned Late Materialization for Ultra-Long Sequence Training in Recommendation Systems at Scale
Versioned late materialization stores user histories once and reconstructs sequences just-in-time during training to cut redundancy and enable longer sequences in large-scale recommendation systems.
-
CS3: Efficient Online Capability Synergy for Two-Tower Recommendation
CS3 strengthens two-tower retrievers via cycle-adaptive feature denoising, cross-tower mutual awareness, and cascade knowledge reuse, delivering consistent gains on public datasets and up to 8.36% revenue lift in prod...
-
MixFormer: Co-Scaling Up Dense and Sequence in Industrial Recommenders
MixFormer unifies dense feature interaction and user-sequence modeling in a single Transformer-style backbone with a user-item decoupling speedup, reporting accuracy and efficiency gains over stacked and parallel reco...
-
Interpretable Representation via LLM-Driven Generative Disentanglement for Local-Life Service Recommendation
Disentangling LLM hidden states into attribute-aligned geo and semantic slots before dual-stream residual quantization cuts SID collisions and improves local-life recommendation AUC.
-
UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
UxSID introduces semantic-group shared interest memory with Semantic IDs and dual-level attention to model ultra-long user sequences, claiming state-of-the-art results and a 0.337% revenue lift in advertising A/B tests.
-
UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
UxSID models ultra-long user sequences with semantic-group shared interest memory using Semantic IDs and dual-level attention, achieving state-of-the-art performance and a 0.337% revenue lift in advertising A/B tests.
-
Versioned Late Materialization for Ultra-Long Sequence Training in Recommendation Systems at Scale
Introduces versioned late materialization to eliminate data redundancy in ultra-long sequence training for DLRMs by storing histories once and reconstructing via pointers at training time.
-
Sample Is Feature: Beyond Item-Level, Toward Sample-Level Tokens for Unified Large Recommender Models
SIF encodes entire historical raw samples as tokens via hierarchical group-adaptive quantization and token/sample-level mixing to overcome partial encoding and feature heterogeneity limits in scaled recommender models.
-
Decoupled Multimodal Fusion for User Interest Modeling in Click-Through Rate Prediction
DMF adds target-aware bridging features and an inference-optimized decoupled attention layer to combine modality-centric and modality-enriched user interest modeling for CTR prediction.
-
MARS: Modality-Aligned Retrieval for Sequence Augmented CTR Prediction
MARS aligns text and image features of items, then augments sparse user sequences with retrieved sequences from similar heavy users, improving CTR prediction offline and online.
-
Diffusion-based Multi-modal Synergy Interest Network for Click-through Rate Prediction
Diff-MSIN improves multi-modal CTR prediction AUC by roughly 1.5-2% over prior multi-modal baselines by disentangling common, specific, and synergistic modality features.
-
SIREN: Unified Multi-Granularity Semantic Interaction for Multi-Modal Lifelong User Interest Modeling
SIREN unifies multi-modal and collaborative features for lifelong user interest modeling via semantic ID retrieval and target-aware transformer interactions, reporting SOTA GAUC and positive GMV gains in production.
-
Sample Is Feature: Beyond Item-Level, Toward Sample-Level Tokens for Unified Large Recommender Models
SIF replaces item-ID history tokens with lossily compressed full-sample tokens and reports consistent CTR/CVR gains in offline and live recommender tests.
-
Deep Situation-Aware Interaction Network for Click-Through Rate Prediction
DSAIN introduces situational features and tri-directional fusion to enhance behavior sequence modeling for CTR prediction, delivering 2.7% CTR, 2.62% CPM, and 2.16% GMV lifts in online A/B tests on the Meituan platform.
-
Coarse-to-Fine Long-term Interest Modeling for Generative Recommendation
GLASS extends generative retrieval with a tiered long-term interest vector and a first-SID-keyed search of long histories, reporting consistent gains over Tiger and DualGR on two public datasets.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.