arXiv subjects

Hanwei Li

Publications and source records attributed to Hanwei Li.

6 recordsLinked to original sources

arXiv2026

Designing Service Systems from Textual Evidence

Designing service systems requires selecting among alternative configurations -- choosing the best chatbot variant, the optimal routing policy, or the most effective quality control procedure. In many service systems, the primary evidence of performance quality is textual -- customer support transcripts, complaint narratives, compliance review reports -- rather than the scalar measurements assumed by classical optimization methods. Large language models (LLMs) can read such textual evidence and produce standardized quality scores, but these automated judges exhibit systematic biases that vary across alternatives and evaluation instances. Human expert review remains accurate but costly. We study how to identify the best service configuration with high confidence while minimizing expensive human audits, given that automated evaluation is cheap but biased. We formalize this as a sequential decision problem where a biased proxy score is observed for every evaluation, and a verified outcome can be acquired selectively at additional cost. We prove that LLM-only selection fails under arm-dependent bias, and that naive selective-audit estimators can be asymptotically biased. We develop an estimator combining proxy scores with inverse-propensity-weighted residuals and construct anytime-valid confidence sequences. Our algorithm, PP-LUCB, jointly decides which alternatives to evaluate and whether to request human audits, concentrating reviews where the LLM judge is least reliable. We prove correctness and establish instance-dependent cost bounds showing near-optimal efficiency. On a customer support ticket classification task, our algorithm correctly identifies the best model in 40/40 trials while achieving 90\% audit cost reduction.

cs.LG↗

arXiv2026

Best Arm Identification with LLM Judges and Limited Human

We study fixed-confidence best-arm identification (BAI) where a cheap but potentially biased proxy (e.g., LLM judge) is available for every sample, while an expensive ground-truth label can only be acquired selectively when using a human for auditing. Unlike classical multi-fidelity BAI, the proxy is biased (arm- and context-dependent) and ground truth is selectively observed. Consequently, standard multi-fidelity methods can mis-select the best arm, and uniform auditing, though accurate, wastes scarce resources and is inefficient. We prove that without bias correction and propensity adjustment, mis-selection probability may not vanish (even with unlimited proxy data). We then develop an estimator for the mean of each arm that combines proxy scores with inverse-propensity-weighted residuals and form anytime-valid confidence sequences for that estimator. Based on the estimator and confidence sequence, we propose an algorithm that adaptively selects and audits arms. The algorithm concentrates audits on unreliable contexts and close arms and we prove that a plug-in Neyman rule achieves near-oracle audit efficiency. Numerical experiments confirm the theoretical guarantees and demonstrate the superior empirical performance of the proposed algorithm.

cs.LG↗

arXiv2025

Quadratic Interest Network for Multimodal Click-Through Rate Prediction

Multimodal click-through rate (CTR) prediction is a key technique in industrial recommender systems. It leverages heterogeneous modalities such as text, images, and behavioral logs to capture high-order feature interactions between users and items, thereby enhancing the system's understanding of user interests and its ability to predict click behavior. The primary challenge in this field lies in effectively utilizing the rich semantic information from multiple modalities while satisfying the low-latency requirements of online inference in real-world applications. To foster progress in this area, the Multimodal CTR Prediction Challenge Track of the WWW 2025 EReL@MIR Workshop formulates the problem into two tasks: (1) Task 1 of Multimodal Item Embedding: this task aims to explore multimodal information extraction and item representation learning methods that enhance recommendation tasks; and (2) Task 2 of Multimodal CTR Prediction: this task aims to explore what multimodal recommendation model can effectively leverage multimodal embedding features and achieve better performance. In this paper, we propose a novel model for Task 2, named Quadratic Interest Network (QIN) for Multimodal CTR Prediction. Specifically, QIN employs adaptive sparse target attention to extract multimodal user behavior features, and leverages Quadratic Neural Networks to capture high-order feature interactions. As a result, QIN achieved an AUC of 0.9798 on the leaderboard and ranked second in the competition. The model code, training logs, hyperparameter configurations, and checkpoints are available at https://github.com/salmon1802/QIN.

cs.IR↗

arXiv2024

FCN: Fusing Exponential and Linear Cross Network for Click-Through Rate Prediction

As an important modeling paradigm in click-through rate (CTR) prediction, the Deep & Cross Network (DCN) and its derivative models have gained widespread recognition primarily due to their success in a trade-off between computational cost and performance. This paradigm employs a cross network to explicitly model feature interactions with linear growth, while leveraging deep neural networks (DNN) to implicitly capture higher-order feature interactions. However, these models still face several key limitations: (1) The performance of existing explicit feature interaction methods lags behind that of implicit DNN, resulting in overall model performance being dominated by the DNN; (2) While these models claim to capture high-order feature interactions, they often overlook potential noise within these interactions; (3) The learning process for different interaction network branches lacks appropriate supervision signals; and (4) The high-order feature interactions captured by these models are often implicit and non-interpretable due to their reliance on DNN. To address the identified limitations, this paper proposes a novel model, called Fusing Cross Network (FCN), along with two sub-networks: Linear Cross Network (LCN) and Exponential Cross Network (ECN). FCN explicitly captures feature interactions with both linear and exponential growth, eliminating the need to rely on implicit DNN. Moreover, we introduce the Self-Mask operation to filter noise layer by layer and reduce the number of parameters in the cross network by half. To effectively train these two cross networks, we propose a simple yet effective loss function called Tri-BCE, which provides tailored supervision signals for each network. We evaluate the effectiveness, efficiency, and interpretability of FCN on six benchmark datasets. Furthermore, by integrating LCN and ECN, FCN achieves a new state-of-the-art performance.

cs.IR↗

arXiv2024

Data-Driven Real-time Coupon Allocation in the Online Platform

Traditionally, firms have offered coupons to customer groups at predetermined discount rates. However, advancements in machine learning and the availability of abundant customer data now enable platforms to provide real-time customized coupons to individuals. In this study, we partner with Meituan, a leading shopping platform, to develop a real-time, end-to-end coupon allocation system that is fast and effective in stimulating demand while adhering to marketing budgets when faced with uncertain traffic from a diverse customer base. Leveraging comprehensive customer and product features, we estimate Conversion Rates (CVR) under various coupon values and employ isotonic regression to ensure the monotonicity of predicted CVRs with respect to coupon value. Using calibrated CVR predictions as input, we propose a Lagrangian Dual-based algorithm that efficiently determines optimal coupon values for each arriving customer within 50 milliseconds. We theoretically and numerically investigate the model performance under parameter misspecifications and apply a control loop to adapt to real-time updated information, thereby better adhering to the marketing budget. Finally, we demonstrate through large-scale field experiments and observational data that our proposed coupon allocation algorithm outperforms traditional approaches in terms of both higher conversion rates and increased revenue. As of May 2024, Meituan has implemented our framework to distribute coupons to over 100 million users across more than 110 major cities in China, resulting in an additional CNY 8 million in annual profit. We demonstrate how to integrate a machine learning prediction model for estimating customer CVR, a Lagrangian Dual-based coupon value optimizer, and a control system to achieve real-time coupon delivery while dynamically adapting to random customer arrival patterns.

econ.EM↗

arXiv2022

Rayleigh-Taylor instability under multi-mode perturbation: discrete Boltzmann modeling with tracers

The Rayleigh-Taylor Instability (RTI) under multi-mode perturbation in compressible flow is probed via the Discrete Boltzmann Modeling (DBM) with tracers. The distribution of tracers provides clear boundaries between light and heavy fluids in the position space. Besides, the position-velocity phase space offers a new perspective for understanding the flow behavior of RTI with intuitive geometrical correspondence. The effects of viscosity, acceleration, compressibility, and Atwood number on the mixing of material and momentum and the mean non-equilibrium strength at the interfaces are investigated separately based on both the mixedness defined by the tracers and the non-equilibrium strength defined by the DBM. The mixedness increases with viscosity during early stage but decreases with viscosity at the later stage. Acceleration, compressibility, and Atwood number show enhancement effects on mixing based on different mechanisms. After the system relaxes from the initial state, the mean non-equilibrium strength at the interfaces presents an initially increasing and then declining trend, which is jointly determined by the interface length and the macroscopic physical quantity gradient. We conclude that the four factors investigated all significantly affect early evolution behavior of RTI system, such as the competition between interface length and macroscopic physical quantity gradient. The results contribute to the understanding of the multi-mode RTI evolutionary mechanism and the accompanied kinetic effects.

physics.flu-dyn↗