论文检索

Giordano Paoletti, Flavio Giobergia, Danilo Giordano, Luca Cagliero, Silvia Ronchiadin, Dario Moncalvo, Marco Mellia, Elena Baralis

This paper presents a real-world deployment case study on using unsupervised anomaly detection for Anti-Money Laundering (AML). Using more than 2 billion anonymized bank transactions that Intesa Sanpaolo, a primary Italian financial institution, registered over 8 months, we developed, tuned and deployed a machine learning pipeline in production. Experts from Intesa Sanpaolo validated the performance of our approach against the institution's traditional rule-based system and checked new real-world cases the system allowed them to identify. Besides increasing both precision and recall by a factor of 6 in the detection of high-risk cases, our pipeline raises 200+ additional alerts during the 8-month period, manually identified by branch managers, but missed by the rule-based system. More importantly, a manual inspection of 100 new unseen cases revealed 28 significant previously unreported cases. The pipeline, now fully deployed in Intesa Sanpaolo's Transaction Monitoring system, highlights the advantages of machine learning over traditional approaches typically adopted in this traditionally very conservative sector.

Bo Pan 0009, Shiva Ghaemi, Amanda J. Consylman, Zihao Zhao, Ashley Ann Petersen, Alice Wu, Gabriel Chang, Diana McDonough, Mark A. Forman, Elise L. Bezold 等

This paper presents a novel Chemist-AI Alignment framework for generating novel structures of quaternary ammonium compounds (QACs), a crucial class of antimicrobial agents. The framework uniquely integrates AI-driven small molecule generation with iterative feedback from chemist experts, leveraging both rapid assessments and comprehensive wet-lab validations to optimize for biological potency and synthetic feasibility. Central to the framework is a hierarchical generative model that captures the QAC hierarchical topology. Extensive experiments highlight the efficacy of the framework in identifying promising QAC candidates, many of which are synthesized and evaluated in the wet laboratory, underscoring the potential of synergistic AI and domain-expert feedback in accelerating the design of novel antimicrobial agents.

Yansong Ning, Shuowei Cai, Wei Li 0176, Jun Fang, Naiqiang Tan, Hua Chai, Hao Liu 0026

On-demand ride-hailing services like DiDi, Uber, and Lyft have transformed urban transportation, offering unmatched convenience and flexibility. In this paper, we introduce DiMA, an LLM-powered ride-hailing assistant deployed in DiDi Chuxing. Its goal is to provide seamless ride-hailing services and beyond through a natural and efficient conversational interface under dynamic and complex spatiotemporal urban contexts. To achieve this, we propose a spatiotemporal-aware order planning module that leverages external tools for precise spatiotemporal reasoning and progressive order planning. Additionally, we develop a cost-effective dialogue system that integrates multi-type dialog repliers with cost-aware LLM configurations to handle diverse conversation goals and trade-off response quality and latency. Furthermore, we introduce a continual fine-tuning scheme that utilizes real-world interactions and simulated dialogues to align the assistant's behavior with human prefered decision-making processes. Since its deployment in the DiDi application, DiMA has demonstrated exceptional performance, achieving 93% accuracy in order planning and 92% in response generation during real-world interactions. Offline experiments further validate DiMA's capabilities, showing improvements of up to 70.23% in order planning and 321.27% in response generation compared to three state-of-the-art agent frameworks, while reducing latency by 0.72x to 5.47x. These results establish DiMA as an effective, efficient, and intelligent mobile assistant for ride-hailing services.

Wenhan Lyu, Devashish Tyagi, Yihang Yang, Ziwei Li, Ajay Somani, Karthikeyan Shanmugasundaram, Nikola Andrejevic, Ferdi Adeputra, Curtis Zeng, Arun K. Singh 等

Long user history is highly valuable signal for recommendation systems, but effectively incorporating it often comes with high cost in terms of data center power consumption and GPU. In this work, we chose offline embedding over end-to-end sequence length optimization methods to enable extremely long user sequence modeling as a cost-effective solution, and propose a new user embedding learning strategy, multi-slicing and summarization, that generates highly generalizable user representation of user's long-term stable interest. History length we encoded in this embedding is up to 70,000 and on average 40,000. This embedding, named as DV365, is proven highly incremental on top of advanced attentive user sequence models deployed in Instagram. Produced by a single upstream foundational model, it is launched in 15 different models across Instagram and Threads with significant impact, and has been production battle-proven for >1 year since our first launch.

Tengfei Lyu, Jindong Han, Hao Liu 0026

Nuclear radiation, which refers to the energy emitted from atomic nuclei during decay, poses significant risks to human health and environmental safety. Recently, advancements in monitoring technology have facilitated the effective recording of nuclear radiation levels and related factors, such as weather conditions. The abundance of monitoring data enables the development of accurate and reliable nuclear radiation forecasting models, which play a crucial role in informing decision-making for individuals and governments. However, this task is challenging due to the imbalanced distribution of monitoring stations over a wide spatial range and the non-stationary radiation variation patterns. In this study, we introduce NRFormer, a novel framework tailored for the nationwide prediction of nuclear radiation variations. By integrating a non-stationary temporal attention module, an imbalance-aware spatial attention module, and a radiation propagation prompting module, NRFormer collectively captures complex spatio-temporal dynamics of nuclear radiation. Extensive experiments on two real-world datasets demonstrate the superiority of our proposed framework against 11 baselines. NRFormer has been deployed online to provide 1-24-day nuclear radiation forecasts, empowering individuals and governments with timely, data-driven decisions for emergency response and public safety. Our framework is designed for general applicability and can be readily adapted for deployment in other regions. The deployed system is available at https://NRFormer.github.io and the dataset and code of the predictive model are available at https://github.com/usail-hkust/NRFormer.

Fuyuan Lyu, Linfeng Du, Yunpeng Weng, Qiufang Ying, Zhiyan Xu, Wen Zou, Haolun Wu, Xiuqiang He 0001, Xing Tang 0007

Fund allocation has been an increasingly important problem in the financial domain. In reality, we aim to allocate the funds to buy certain assets within a certain future period. Naive solutions such as prediction-only or Predict-then-Optimize approaches suffer from goal mismatch. Additionally, the introduction of the SOTA time series forecasting model inevitably introduces additional uncertainty in the predicted result. To solve both problems mentioned above, we introduce a Risk-aware Time-Series Predict-and-Allocate (RTS-PnO) framework, which holds no prior assumption on the forecasting models. Such a framework contains three features: (i) end-to-end training with objective alignment measurement(ii) adaptive forecasting uncertainty calibration, and (iii) agnostic towards forecasting models. The evaluation of RTS-PnO is conducted over both online and offline experiments. For offline experiments, eight datasets from three categories of financial applications are used: Currency, Stock, and Cryptos. RTS-PnO consistently outperforms other competitive baselines. The online experiment is conducted on the Cross-Border Payment business at FiT, Tencent, and an 8.4% decrease in regret is witnessed when compared with the product-line approach. The code for the offline experiment is available.

Xingyu Lu, Tianke Zhang, Chang Meng, Xiaobei Wang, Jinpeng Wang 0002, Yifan Zhang 0004, Shisong Tang, Changyi Liu, Haojie Ding, Kaiyu Jiang 等

Exponentially growing short video platforms (SVPs) face significant challenges in moderating content detrimental to users' mental health, particularly for minors. The dissemination of such content on SVPs can lead to catastrophic societal consequences. Although substantial efforts have been dedicated to moderating such content, existing methods suffer from critical limitations: (1) Manual review is prone to human bias and incurs high operational costs. (2) Automated methods, though efficient, lack nuanced content understanding, resulting in lower accuracy. (3) Industrial moderation regulations struggle to adapt to rapidly evolving trends due to long update cycles. In this paper, we annotate the first SVP content moderation benchmark with authentic user/reviewer feedback to fill the absence of benchmark in this field. Then we evaluate various methods on the benchmark to verify the existence of the aforementioned limitations. We further propose our common-law content moderation framework named KuaiMod to address these challenges. KuaiMod consists of three components: training data construction, offline adaptation, and online deployment & refinement. Leveraging large vision language model (VLM) and Chain-of-Thought (CoT) reasoning, KuaiMod adequately models video toxicity based on sparse user feedback and fosters dynamic moderation policy with rapid update speed and high accuracy. Offline experiments and large-scale online A/B test demonstrates the superiority of KuaiMod: KuaiMod achieves the best moderation performance on our benchmark. The deployment of KuaiMod reduces the user reporting rate by 20% and its application in video recommendation increases both Daily Active User (DAU) and APP Usage Time (AUT) on several Kuaishou scenarios. We have open-sourced our benchmark at https://kuaimod.github.io.

Xingyu Lu, Jinpeng Wang 0002, Jieming Zhu, Zhicheng Zhang 0008, Deqing Zou, Hai-Tao Zheng 0002, Shu-Tao Xia, Rui Zhang 0003

Sequential recommendation (SR) aims to capture dynamic user preferences from users' historical behaviors. Recently, benefiting from astonishing understanding ability of pre-trained language models (PLMs), text-enhanced sequential recommender becomes a promising direction, which employs PLMs to extract semantic information for user/item representation. Although promising in improving performance and transferability, few existing text-enhanced SR studies have analyzed the differences between PLMs and recommenders, restricting the ability of PLMs for recommendation. In this paper, we make an in-depth comparison and conclude their discrepancies in representation and knowledge level, respectively, caused by different multi-modal content and task-oriented capabilities. Based on this, we propose a Recommendation-Oriented Language Model Adaptation framework (named ROMA) using multi-modal multi-domain item sequences. To empower PLMs with a rational understanding of user/item modeling and the recommendation task, ROMA partitions a PLM into bottom and top layers, respectively, allowing representation-level and task-level adaptation with elaborately designed architectures, transferring strategy and learning framework. Our experimental results on public benchmarks demonstrate the effectiveness and transferability of our framework. Additionally, we showcase the application value of ROMA on the recommender system of Huawei's AppGallery through online A/B testing, which shows significant improvements in online metrics.

Ping Liu 0002, Rajat Arora 0002, Xiao Shi, Benjamin Le, Qianqi Shen, Jianqiang Shen, Chengming Jiang 0001, Nikita Zhiltsov, Priya Bannur, Yidan Zhu 等

LinkedIn, one of the world's largest platforms for professional networking and job seeking, encounters various modeling challenges in building recommendation systems for its job matching product, including cold-start, filter bubbles, and biases affecting candidate-job matching. To address these, we developed the STAR (Signal integration for Talent And Recruiters) system, leveraging the combined strengths of Large Language Models (LLMs) and Graph Neural Networks (GNNs). LLMs excel at understanding textual data, such as member profiles and job postings, while GNNs capture intricate relationships and mitigate cold-start issues through network effects. STAR integrates diverse signals by uniting LLM and GNN capabilities with industrial-scale paradigms including adaptive sampling and version management. It provides an end-to-end solution for developing and deploying embeddings in large-scale recommender systems. Our key contributions include a robust methodology for building embeddings in industrial applications, a scalable GNN-LLM integration for high-performing recommendations, and practical insights for real-world model deployment.

Ning Liu 0014, Yunsen Tang, Haitao Yuan 0002, Hongtao Lv, Lili Jiang 0002, Zhen Li 0049, Wei Zhang 0056, Jianyong Wang 0001

The drug recommendation task aims to provide effective and safe prescription decision support for clinical treatment based on patients' past Electronic Health Records (EHR). However, the prevalent phenomenon of missing views in multi-source heterogeneous EHR data may cause performance degradation. This is due to the lack of sufficient information and increased learning difficulties, which limit the practical effectiveness of drug recommendation models in medical applications. In this paper, we emphasize the problems of incompleteness in practical drug recommendation and propose the Incomplete Multi-View Drug Recommendation model via Multi-Level Representation Learning and Curriculum Learning named IMDR. In particular, IMDR employs a Multi-Level Representation Learning architecture equipped with a Medical Code-Level Drug Knowledge Infusion Module and a Visit-Level Cross-View Information Module for patient representation learning to overcome the information loss caused by incomplete data. And then, a Gaussian-guided curriculum learning strategy is proposed to assist the learning process of IMDR with a novel difficulty measure to achieve effective progressive learning under missing medical views. Systematic evaluation on two large-scale real-world medical datasets, MIMIC-III and MIMIC-IV, demonstrates that IMDR reduces the Drug-Drug Interaction (DDI) rate by 2.97% compared to existing state-of-the-art drug recommendation baselines, while achieving significant improvements of 3.29% and 1.97% in Jaccard similarity scores and F1 score, respectively. Furthermore, compared to advanced incomplete multi-view learning (IML) models, IMDR's advantages in Jaccard similarity scores and F1 score further expand to 4.03% and 2.41%.

Langming Liu, Wanyu Wang, Chi Zhang 0060, Bo Li 0156, Hongzhi Yin, Xuetao Wei, Wenbo Su, Bo Zheng 0007, Xiangyu Zhao 0001

Online advertising in recommendation platforms has gained significant attention, with a predominant focus on channel recommendation and budget allocation strategies. However, current offline reinforcement learning (RL) methods face substantial challenges when applied to sparse advertising scenarios, primarily due to severe overestimation, distributional shifts, and overlooking budget constraints. To address these issues, we propose MTORL, a novel multi-task offline RL model that targets two key objectives. First, we establish a Markov Decision Process (MDP) framework specific to the nuances of advertising. Then, we develop a causal state encoder to capture dynamic user interests and temporal dependencies, facilitating offline RL through conditional sequence modeling. Causal attention mechanisms are introduced to enhance user sequence representations by identifying correlations among causal states. We employ multi-task learning to decode actions and rewards, simultaneously addressing channel recommendation and budget allocation. Notably, our framework includes an automated system for integrating these tasks into online advertising. Extensive experiments on offline and online environments demonstrate MTORL's superiority over state-of-the-art methods. The code is available online at https://github.com/Applied-Machine-Learning-Lab/MTORL.

Jun Liu, Chaoyun Zhang, Jiaxu Qian, Minghua Ma, Si Qin, Chetan Bansal, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang 0001

Time series anomaly detection (TSAD) plays a crucial role in various industrial applications. Traditional deep learning TSAD models require extensive training data and operate as black boxes, lacking interpretability for detected anomalies. To address these challenges, we propose LLMAD, a novel TSAD method that employs Large Language Models (LLMs) to deliver accurate and interpretable TSAD results. LLMAD applies in-context anomaly detection by retrieving both positive and negative similar time series segments, significantly enhancing LLMs' effectiveness. Furthermore, LLMAD employs the Anomaly Detection Chain-of-Thought approach to mimic expert logic for its decision-making process. This further enhances its performance and enables LLMAD to provide explanations for their detections through versatile perspectives. Experiments conducted on both offline datasets and through online deployment on a large-scale cloud system at Microsoft indicate that our LLMAD achieves detection performance comparable to state-of-the-art deep learning methods. Additionally, LLMAD offers human-readable interpretability for detections at a reasonable cost, significantly reducing engineers' effort.

Jingxin Liu 0003, Xiang Gao 0039, YiSha Li 0001, Xin Li 0243, Haiyang Lu 0001, Ben Wang 0006

In the context of a short video & live stream mixed recommendation scenario, the live stream recommendation system (RS) decides whether to allocate at most one live stream to the video feed for each user request. The inappropriate policy which ignores the long-term negative impact of live stream allocation can significantly affect app usage duration and user retention. To maximize long-term user engagement, it is crucial to determine an optimal policy for accurate live stream allocation. Recently, reinforcement learning (RL) has been widely applied in recommendation systems to capture long-term user engagement. However, traditional RL algorithms often face divergence and instability problems, which restricts application and deployment in large-scale industrial recommendation systems, especially in the aforementioned challenging scenario. To address these challenges, we propose a novel Supervised Learning-enhanced Multi-Group Actor Critic algorithm (SL-MGAC). Specifically, we introduce a supervised learning-enhanced actor-critic framework that incorporates variance reduction techniques, where multi-task supervised reward learning helps restrict bootstrapping error accumulation during critic learning. Additionally, we design a multi-group state decomposition module for both actor and critic networks to reduce prediction variance and improve model stability. We also propose a novel reward function to prevent overly greedy live stream allocation. Empirically, we evaluate the SL-MGAC algorithm using offline policy evaluation (OPE) and online A/B testing. Experimental results demonstrate that the proposed method not only outperforms baseline methods under platform-level constraints, but also exhibits improved stability in online recommendation scenarios.

Hongrui Liu, Daixin Wang, Yifan Wu 0020, Zhiqiang Zhang 0012, Xiaobo Qin

As the demand for insurance continues to skyrocket in our daily lives, accurately predicting claim amount has become a critical demand for insurance companies. This capacity enables to identify high-risk individuals for minimizing substantial claims and, conversely, help to lower premiums for other policyholders. Recently, graph neural networks (GNNs) have achieved remarkable performance on classification and recommendation for insurance, but have not been extended to this regression task. Will GNNs serve as a powerful tool for claim amount prediction? Surprisingly, our research reveals that GNNs perform well in this area, but encounter three challenges, i.e., the mixture of heterophilic and homophilous network pattern, label scarcity and label uncertainty. To address these challenges, we propose Decoupled Graph Neural Networks Enhanced by Pseudo Labels for Claim Prediction (GClaim) to empower existing GNNs for claim amount prediction. Specifically, GClaim automatically organizing nodes into distinct clusters, facilitating the independent learning of nodes within each latent network pattern. It further introduces trustworthy pseudo labels from unlabeled and uncertain nodes through a standard deviation-induced filtering strategy. Extensive experiments on three industrial datasets and five newly developed proxy public datasets with varying evaluation protocols demonstrate the effectiveness of GClaim, as well as each well-designed component in GClaim. In light of its outstanding performance, GClaim has been successfully deployed in the online auto-insurance platform of Alipay. With GClaim, the auto-insurance service has experienced an over 10% increase in the end-to-end conversion rate and over 25% rise in UV value.

Hongyu Lin, Shuxin Zhong, Yan Fang, Zhiqing Hong, Wenjun Lyu, Qipeng Xie, Haotian Wang 0008, Lu Wang 0002, Kaishun Wu

An emerging type of fraud involves malicious senders exploiting the blind shipment and cash-on-delivery (COD) mechanisms by dispatching large volumes of unsolicited, low-cost parcels. If unsuspecting receivers accept these parcels, they pay for both shipping and goods; otherwise, logistics providers bear the round-trip shipping costs. Existing detection techniques, which rely on extensive labeled cases, struggle with this emerging fraud because receivers' unawareness and low transaction values discourage complaints, resulting in few confirmed cases. Therefore, we propose leveraging receivers' complaints, though not initially collected for fraud detection, to uncover subtle indicators of fraud patterns, while addressing three challenges: (C1) noise-rich dialogues(C2) data privacy concerns, and (C3) ever-evolving fraud patterns. To address them, we design BLOFF, a Blind shipment detection Framework for LO gistics Fraud powered by large language models (LLMs). Specifically, BLOFF includes three components: i) Sensitivity Anonymization to protect sensitive user information; ii) Dialogue Profile Distillation to transform informal dialogues into structured representation, addressing C1, and distill knowledge from a teacher LLM (GPT-4o) to a lightweight student LLM (ChatGLM4-9B), addressing C2; ii) Multi-faceted Context Augmentation to enhance the interpretation of fraud signatures and adaptation of evolving patterns, addressing C3. We evaluate BLOFF on about 56,000 complaints records collected from JD Logistics between January and November 2024. Results show that BLOFF outperforms state-of-the-art methods, achieving a 10.19% improvement in precision. Furthermore, during its real-world deployment in December 2024, BLOFF identified over 90 fraudulent parcels with a 91.4% precision.

Yuchen Li 0006, Hao Zhang 0156, Yongqi Zhang, Hengyi Cai, Mingxin Cai, Shuaiqiang Wang, Haoyi Xiong, Linghe Kong, Dawei Yin 0001, Lei Chen 0002

As modern learning-to-rank (LTR) systems rely on both textual and behavioral features, it is essential to extend pre-trained language models (PLMs) from text (queries and webpages) understanding to end-to-end ranking score prediction subject to multiple objectives, such as relevance, quality, authority, and recency. While textual inputs encompass a broader array of features than mere relevance and behavioral features are frequently skewed by user feedback with position bias, an integrated solution is required to jointly disentangle and fuse these heterogeneous features, ensuring robust and unbiased ranking predictions. In this work, we introduce RankExpert, a unified framework that holistically models heterogeneous ranking signals by integrating PLM-based semantic extraction with behavioral cues. RankExpert employs a lightweight PLM with hierarchical distillation for efficient query-document representation, a mixture-of-experts (MoE) architecture to disentangle and optimize objectives across relevance, quality, authority, and recency, and a click expert to mitigate position bias in user feedback within a dual-tower design. Then, RankExpert utilizes an adaptive weight fusion layer to dynamically integrate the outputs of these specialized experts dynamically, aligning ranking predictions with diverse user intents. Extensive offline evaluations on two large-scale real-world datasets reveal that RankExpert significantly outperforms strong competitor models and substantial gains in key performance indicators. Moreover, we deploy RankExpert at Baidu Search and conduct comprehensive online evaluations with real-world web traffic, demonstrating substantial improvements in user satisfaction metrics over the online legacy system and underscoring the practicality of the proposed approach.

Xiaochong Lan, Jie Feng 0002, Jiahuan Lei, Xinlei Shi, Yong Li 0008

Large language models (LLMs) have exhibited remarkable capabilities and achieved significant breakthroughs across various domains, leading to their widespread adoption in recent years. Building on this progress, we investigate their potential in the realm of local life services. In this study, we establish a comprehensive benchmark and systematically evaluate the performance of diverse LLMs across a wide range of tasks relevant to local life services. To further enhance their effectiveness, we explore two key approaches: model fine-tuning and agent-based workflows. Our findings reveal that even a relatively compact 7B model can attain performance levels comparable to a much larger 72B model, effectively balancing inference cost and model capability. This optimization greatly enhances the feasibility and efficiency of deploying LLMs in real-world online services, making them more practical and accessible for local life applications. Available resources are at https://github.com/tsinghua-fib-lab/LocalEval.

Hao Jiang 0051, Xiaoyu He, Fanyi Qu, Congcong Liu, Xue Jiang, Changping Peng, Zhangang Lin, Ching Law, Jingping Shao

E-commerce has become an integral part of daily life, and the ability to effectively retrieve items relevant to a user's query is crucial for enhancing the shopping experience. Embedding-based retrieval (EBR) has proven to be an effective approach in industrial e-commerce search systems. This method involves training models to generate high-quality representations of queries and items, followed by the use of efficient approximate nearest neighbor (ANN) search techniques to find relevant items. However, current EBR methods face several critical limitations: (1) multiple EBR branches often retrieve overlapping or redundant result sets; (2) allocation and utilization of computational resources remain suboptimal, hindering performance; (3) the differentiation modeling of features is somewhat neglected, which restricts the system's ability to retrieve diverse and representative results. These issues hinder the retrieval performance of online search systems. In this paper, we introduce a novel e-commerce search framework called the Uniform Embedding-based Retrieval Framework (UniERF). This framework is meticulously designed to incorporate diverse samples for joint model training, enabling the model to effectively leverage both the semantic information of queries and the personalized features of different users. Extensive offline and online experiments demonstrate that UniERF surpasses baseline methods across various evaluation metrics. UniERF has been successfully implemented in the existing retrieval system at JD.COM, a renowned online shopping website.

Jiahao Ji, Tianyu Wang 0028, Yeshu Li, Yusen Huo, Zhilin Zhang 0003, Chuan Yu 0002, Jian Xu 0015, Bo Zheng 0007

Auto-bidding is crucial in facilitating online advertising by automatically providing bids for advertisers. While previous work has made great efforts to model bidding environments for better ad performance, it has limitations in generalizability across environments since these models are typically tailored for specific bidding scenarios. To this end, we approach the scenario-independent principles through a unified function that estimates the achieved effect under specific bids, such as budget consumption, gross merchandise volume (GMV), page views, etc. Then, we propose a bidding foundation model Bid2X to learn this fundamental function from data in various scenarios. Our Bid2X is built over uniform series embeddings that encode heterogeneous data through tailored embedding methods. To capture complex inter-variable and dynamic temporal dependencies in bidding data, we propose two attention mechanisms separately treating embeddings of different variables and embeddings at different times as attention tokens for representation learning. On top of the learned variable and temporal representations, a variable-aware fusion module is used to perform adaptive bidding outcome prediction. To model the unique bidding data distribution, we devise a zero-inflated projection module to incorporate the estimated non-zero probability into its value prediction, which makes up a joint optimization objective containing classification and regression. The objective is proven to converge to the zero-inflated distribution. Our model has been deployed on the ad platform in Taobao, one of the world's largest e-commerce platforms. Offline evaluation on eight large-scale real-world datasets exhibits Bid2X's superiority compared to various baselines and its generality across different scenarios. In real-world applications, Bid2X increased GMV by 4.65% and ROI by 2.44% in online A/B tests, paving the way for the bidding foundation model in computational advertising.

Maolei Huang, Detao Lv, Yao Yu, Shuhan Song, Dong Li 0037, Zhuoran Zhuang

With the flourishing development of aviation and the convenience of booking flights online, nearby flight recommendation has become the core business of Online Travel Platforms (OTPs). Nearby flight addresses the issue of inadequate flight options for travelers by offering more cost-effective alternatives, such as recommending flights from nearby cities or on nearby departure dates. Currently, mainstream OTPs adopt rule-based or simple user preference-based strategies to recommend nearby flights. However, the insufficient emphasis on the user's historical behaviors and the ignorance of nearby flight's context make these existing strategies less effective in solving the nearby flight recommendation. To this end, a Context-based Personalized Deep Net work (CPNet) is proposed in this paper for nearby flight recommendation. In CPNet, a Personalized Preferences Learning (PPL) component is first proposed to encapsulate users' individual preferences, leveraging crucial feature correlations between historical behaviors and target nearby flight. Then, a Historical Cost Learning (HCL) component is designed to learn the price sensitivity of users under the same query and the same nearby flight recommendation. Finally, we present a Context Potential Gain Learning (CPGL) component, where the important cost between target nearby flight and context flights are emphasized and learned. Offline experiments on a production dataset and a world-scale online A/B test at Fliggy. Fliggy: https://www.fliggy.com/ both demonstrate the superiority of the proposed CPNet over baselines. CPNet is now successfully deployed at Fliggy, one of the largest OTPs in China, serving millions of users every day for flight reservations.