cs.IRFeb 24, 2026

PRECTR-V2:Unified Relevance-CTR Framework with Cross-User Preference Mining, Exposure Bias Correction, and LLM-Distilled Encoder Optimization

Authors: Shuzhi Cao, Rong Chen, Ailong He, Shuguang Han, Jufeng Chen

Organizations: Alibaba Group, Hangzhou, China

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

In search systems, effectively coordinating the two core objectives of search relevance matching and click-through rate (CTR) prediction is crucial for discovering users' interests and enhancing platform revenue. In our prior work PRECTR, we proposed a unified framework to integrate these two subtasks,thereby eliminating their inconsistency and leading to mutual benefit.However, our previous work still faces three main challenges. First, low-active users and new users have limited search behavioral data, making it difficult to achieve effective personalized relevance preference modeling. Second, training data for ranking models predominantly come from high-relevance exposures, creating a distribution mismatch with the broader candidate space in coarse-ranking, leading to generalization bias. Third, due to the latency constraint, the original model employs an Emb+MLP architecture with a frozen BERT encoder, which prevents joint optimization and creates misalignment between representation learning and CTR fine-tuning. To solve these issues, we further reinforce our method and propose PRECTR-V2. Specifically, we mitigate the low-activity users' sparse behavior problem by mining global relevance preferences under the specific query, which facilitates effective personalized relevance modeling for cold-start scenarios. Subsequently, we construct hard negative samples through embedding noise injection and relevance label reconstruction, and optimize their relative ranking against positive samples via pairwise loss, thereby correcting exposure bias. Finally, we pretrain a lightweight transformer-based encoder via knowledge distillation from LLM and SFT on the text relevance classification task. This encoder replaces the frozen BERT module, enabling better adaptation to CTR fine-tuning and advancing beyond the traditional Emb+MLP paradigm.

Figures & tables

Figure 1. The overview of the proposed PRECTR-V2 method.
Figure 2. The overview of the debias process.
Figure 3. (a) exhibits the predicted average score between the click and unclick samples, and (b) exhibits the predicted average score difference between the samples with different relevance score levels.
Figure 4. The overview of the lightweight encoder training.
MethodAUCRIGAUCRI
LR ( Hosmer Jr et al., 2013 )0.6795-29.10%0.6347-26.6%
DNN0.75410.35%0.68631.47%
Wide&Deep ( Cheng et al., 2016 )0.75320.00%0.68360.00%
DeepFM ( Guo et al., 2017 )0.7519-0.51%0.68390.16%
XDeepFM ( Lian et al., 2018 )0.7521-0.43%0.6843-0.29%
DIN ( Zhou et al., 2018 )0.75611.15%0.68752.21%
Table 1. The offline comparison results, where "RI" is short for "RelaImpr".
MethodAUCRIGAUCRI
PRECTR0.75810.00%0.68920.00%
w/o Cold-Start Relevance Preference Mining0.76091.08%0.69080.85%
w/o Relevance Exposure Debias0.76582.98%0.69281.91%
w/o LLM-Distilled and CTR-Aligned Encoder0.76623.13%0.69271.85%
PRECTR-V20.7674\boldsymbol{0.7674}3.60%\boldsymbol{3.60\%}0.6933\boldsymbol{0.6933}2.16%\boldsymbol{2.16\%}
Table 2. The ablation study of different PRECTR-V2 variants on production datasets, where "w/o" is short for "without".
Figure 5. The selection of hyperparameters.
More in this topic

CTR Prediction

Click-through rate (CTR) prediction estimates which ads or items users will click to improve ranking, recommendation, and advertising decisions.

See all 19 papers

Explore similar work

CardsList
Jun 8, 2026

DUET -- Dual User Embedding Transformers for Offsite Conversion Prediction

Offsite conversion rate (OCVR) prediction is an important ranking problem in computational recommendation systems. This task presents a modeling challenge: click signals are abundant and exhibit short temporal horizons, whereas conversion signals are inherently sparse, long-delayed, and frequently unattributed. Despite these statistical disparities, both signal types must inform models that operate within strict serving-latency constraints. Prior pre-training approaches address this heterogeneity with a single, undifferentiated encoder applied uniformly across both data streams. We propose DUET (Dual User Embedding Transformers), a framework that explicitly partitions user behavioral data into two domain-coherent streams -- clicks and conversions -- and pre-trains dedicated transformer encoders with architectures tailored to each stream's statistical characteristics: multi-layer self-attention for the dense click stream and interleaved cross- and self-attention for the sparse conversion stream. The resulting complementary embeddings are jointly consumed by a downstream ranker without exceeding serving-latency budgets. Evaluation demonstrates up to 0.38% normalized entropy (NE) reduction relative to the strongest baseline, and A/B test shows consistent improvements in OCVR prediction accuracy.

Reazul Hasan Russel, Mingwei Tang, Rostam Shirani +18

AI at Meta

Learning to RankUser ModelingPredictive Representation LearningSequence ModelingConversion Rate Prediction
AskPDF
May 15, 2026

Generative Long-term User Interest Modeling for Click-Through Rate Prediction

Modeling long-term user interests with massive historical user behaviors enhances click-through rate (CTR) prediction performance in advertising and recommendation systems. Typically, a two-stage framework is widely adopted, where a general search unit (GSU) first retrieves top-kk relevant behaviors towards the target item, and an exact search unit (ESU) generates interest features via tailored attention. However, current target-centered GSU would ignore other latent user interests, leading to incomplete and biased interest features. Additionally, the matching-based retrieval process in GSUs depends on the pairwise similarity score between target item and each historical behavior, which not only becomes time-consuming for online services as user behaviors continue to grow, but also overlooks the interaction information among user behaviors. To combat these problems, we propose a Generative Long-term user Interest model named GenLI for CTR prediction. GenLI consists of an interest generation module (IGM), a behavior retrieval module (BRM), and an interest fusion module (IFM). The IGM generates multiple interest distributions to indicate different aspects of real-time user interests, which is target-independent and incorporates interaction information among behaviors, ensuring complete and diverse interest features. The BRM selects related behaviors via a simple lookup operation, reducing the time complexity for weighting each behavior to O(1)O(1). Finally, the IFM uses delicate gating mechanisms to generate interest features. Based on the generation process, GenLI improves the diversity of user interests and avoids complex matching-based behavioral retrieval, achieving a better balance between accuracy and efficiency for CTR prediction.

Jiangli Shao, Kaifu Zheng, Hao Fang +5

MeiTuan Beijing, China

CTR PredictionUser ModelingRecommender SystemsUser Preference Modeling
AskPDF
Jun 12, 2026

CoRe: A Continuously Reward-Finetuned LLM Query Rewriter for Multi-Stage Context-Aware Relevance in Web-Scale Video Search

LLM-based query rewriters in production face a tension: the training reward must reflect how the rewrite is consumed by the production ranker, yet the training procedure must be cheap enough to support continuous redeployment as data drifts. We present CoRe (Context Relevance), such a system, redeployed weekly for over five months in a major short-video search engine. Our reward uses the deployed multimodal relevance model as its source and a multiplicative ratio form mirroring the production fusion algebra, closing the simulation-production gap that offline reward proxies leave open. A semi-online Mixed Preference Optimization loop makes this reward affordable at multi-million-instance weekly scale: a DPO-style pairwise objective restricts the gradient pass to a small top-k/bottom-k subset of sampled trajectories, and a phase structure reduces trainer/inference-server parameter syncs from per-step to per-phase. An automated promotion gate over reward-like and stability metrics detected and recovered from a real reward-hacking incident in production. Rewriter output is consumed as parallel relevance signals at recall, rawrank, and finerank without displacing the original signals, bounding rewriter-failure blast radius. Online A/B from two sequential production launches, first deploying the rewriter at finerank, then extending consumption to recall and rawrank, delivers statistically significant reductions in change-query rate on rewrite-impacted queries, with all headline relevance and engagement metrics moving in the expected direction.

Yilin Wen, Rong Yang, Xiaojia Chang +11

TikTok

LLM Fine-Tuning
AskPDF