Hasil untuk "Commerce"

R. Rosecrance

Zhouwei Zhai, Mengxiang Chen, Haoyun Xia et al.

Modern e-commerce search engines, largely rooted in passive retrieval-and-ranking models, frequently fail to support complex decision-making, leaving users overwhelmed by cognitive friction. In this paper, we introduce CogSearch, a novel cognitive-oriented multi-agent framework that reimagines e-commerce search as a proactive decision support system. By synergizing four specialized agents, CogSearch mimics human cognitive workflows: it decomposes intricate user intents, fuses heterogeneous knowledge across internal and external sources, and delivers highly actionable insights. Our offline benchmarks validate CogSearch's excellence in consultative and complex search scenarios. Extensive online A/B testing on JD.com demonstrates the system's transformative impact: it reduced decision costs by 5% and achieved a 0.41% increase in overall UCVR, with a remarkable 30% surge in conversion for decision-heavy queries. CogSearch represents a fundamental shift in information retrieval, moving beyond traditional relevance-centric paradigms toward a future of holistic, collaborative decision intelligence.

Jiaqi Xi, Raghav Saboo, Luming Chen et al.

We propose a two-stage "Mine and Refine" contrastive training framework for semantic text embeddings to enhance multi-category e-commerce search retrieval. Large scale e-commerce search demands embeddings that generalize to long tail, noisy queries while adhering to scalable supervision compatible with product and policy constraints. A practical challenge is that relevance is often graded: users accept substitutes or complements beyond exact matches, and production systems benefit from clear separation of similarity scores across these relevance strata for stable hybrid blending and thresholding. To obtain scalable policy consistent supervision, we fine-tune a lightweight LLM on human annotations under a three-level relevance guideline and further reduce residual noise via engagement driven auditing. In Stage 1, we train a multilingual Siamese two-tower retriever with a label aware supervised contrastive objective that shapes a robust global semantic space. In Stage 2, we mine hard samples via ANN and re-annotate them with the policy aligned LLM, and introduce a multi-class extension of circle loss that explicitly sharpens similarity boundaries between relevance levels, to further refine and enrich the embedding space. Robustness is additionally improved through additive spelling augmentation and synthetic query generation. Extensive offline evaluations and production A/B tests show that our framework improves retrieval relevance and delivers statistically significant gains in engagement and business impact.

Qiaolei Gu, Yu Li, DingYi Zeng et al.

In e-commerce advertising, selecting the most compelling combination of creative elements -- such as titles, images, and highlights -- is critical for capturing user attention and driving conversions. However, existing methods often evaluate creative components individually, failing to navigate the exponentially large search space of possible combinations. To address this challenge, we propose a novel framework named GenCO that integrates generative modeling with multi-instance reward learning. Our unified two-stage architecture first employs a generative model to efficiently produce a diverse set of creative combinations. This generative process is optimized with reinforcement learning, enabling the model to effectively explore and refine its selections. Next, to overcome the challenge of sparse user feedback, a multi-instance learning model attributes combination-level rewards, such as clicks, to the individual creative elements. This allows the reward model to provide a more accurate feedback signal, which in turn guides the generative model toward creating more effective combinations. Deployed on a leading e-commerce platform, our approach has significantly increased advertising revenue, demonstrating its practical value. Additionally, we are releasing a large-scale industrial dataset to facilitate further research in this important domain.

Zipeng Guo, Lichen Ma, Xiaolong Fu et al.

In web data, product images are central to boosting user engagement and advertising efficacy on e-commerce platforms, yet the intrusive elements such as watermarks and promotional text remain major obstacles to delivering clear and appealing product visuals. Although diffusion-based inpainting methods have advanced, they still face challenges in commercial settings due to unreliable object removal and limited domain-specific adaptation. To tackle these challenges, we propose Repainter, a reinforcement learning framework that integrates spatial-matting trajectory refinement with Group Relative Policy Optimization (GRPO). Our approach modulates attention mechanisms to emphasize background context, generating higher-reward samples and reducing unwanted object insertion. We also introduce a composite reward mechanism that balances global, local, and semantic constraints, effectively reducing visual artifacts and reward hacking. Additionally, we contribute EcomPaint-100K, a high-quality, large-scale e-commerce inpainting dataset, and a standardized benchmark EcomPaint-Bench for fair evaluation. Extensive experiments demonstrate that Repainter significantly outperforms state-of-the-art methods, especially in challenging scenes with intricate compositions. We will release our code and weights upon acceptance.

Jungbae Park, Heonseok Jang

E-commerce search optimization has evolved to include a wider range of metrics that reflect user engagement and business objectives. Modern search frameworks now incorporate advanced quality features, such as sales counts and document-query relevance, to better align search results with these goals. Traditional methods typically focus on click-through rate (CTR) as a measure of engagement or relevance, but this can miss true purchase intent, creating a gap between user interest and actual conversions. Joint training with the click-through conversion rate (CTCVR) has become essential for understanding buying behavior, although its sparsity poses challenges for reliable optimization. This study presents MOHPER, a Multi-Objective Hyperparameter Optimization framework for E-commerce Retrieval systems. Utilizing Bayesian optimization and sampling, it jointly optimizes both CTR, CTCVR, and relevant objectives, focusing on engagement and conversion of the users. In addition, to improve the selection of the best configuration from multi-objective optimization, we suggest advanced methods for hyperparameter selection, including a meta-configuration voting strategy and a cumulative training approach that leverages prior optimal configurations, to improve speeds of training and efficiency. Currently deployed in a live setting, our proposed framework substantiates its practical efficacy in achieving a balanced optimization that aligns with both user satisfaction and revenue goals.

Pengkun Jiao, Yiming Jin, Jianhui Yang et al.

Query-product relevance prediction is vital for AI-driven e-commerce, yet current LLM-based approaches face a dilemma: SFT and DPO struggle with long-tail generalization due to coarse supervision, while traditional RLVR suffers from sparse feedback that fails to correct intermediate reasoning errors. We propose Stepwise Hybrid Examination (SHE), an RL framework that ensures logical consistency through Stepwise Reward Policy Optimization (SRPO). SRPO utilizes a hybrid reward mechanism-combining generative reward models with human-annotated verifiers-to provide fine-grained, step-level signals. To further enhance stability, SHE incorporates diversified data filtering to maintain policy entropy and a multi-stage curriculum learning protocol for progressive skill acquisition. Extensive experiments on real-world search benchmarks show that SHE improves both reasoning quality and relevance-prediction accuracy in large-scale e-commerce settings, outperforming SFT, DPO, GRPO, and other baselines, while also enhancing interpretability and robustness.

Divij Handa, David Blincoe, Orson Adams et al.

Deploying capable and user-aligned LLM-based systems necessitates reliable evaluation. While LLMs excel in verifiable tasks like coding and mathematics, where gold-standard solutions are available, adoption remains challenging for subjective tasks that lack a single correct answer. E-commerce Query Rewriting (QR) is one such problem where determining whether a rewritten query properly captures the user intent is extremely difficult to figure out algorithmically. In this work, we introduce OptAgent, a novel framework that combines multi-agent simulations with genetic algorithms to verify and optimize queries for QR. Instead of relying on a static reward model or a single LLM judge, our approach uses multiple LLM-based agents, each acting as a simulated shopping customer, as a dynamic reward signal. The average of these agent-derived scores serves as an effective fitness function for an evolutionary algorithm that iteratively refines the user's initial query. We evaluate OptAgent on a dataset of 1000 real-world e-commerce queries in five different categories, and we observe an average improvement of 21.98% over the original user query and 3.36% over a Best-of-N LLM rewriting baseline.

Weiqi Wang, Limeng Cui, Xin Liu et al.

Goal-oriented script planning, or the ability to devise coherent sequences of actions toward specific goals, is commonly employed by humans to plan for typical activities. In e-commerce, customers increasingly seek LLM-based assistants to generate scripts and recommend products at each step, thereby facilitating convenient and efficient shopping experiences. However, this capability remains underexplored due to several challenges, including the inability of LLMs to simultaneously conduct script planning and product retrieval, difficulties in matching products caused by semantic discrepancies between planned actions and search queries, and a lack of methods and benchmark data for evaluation. In this paper, we step forward by formally defining the task of E-commerce Script Planning (EcomScript) as three sequential subtasks. We propose a novel framework that enables the scalable generation of product-enriched scripts by associating products with each step based on the semantic similarity between the actions and their purchase intentions. By applying our framework to real-world e-commerce data, we construct the very first large-scale EcomScript dataset, EcomScriptBench, which includes 605,229 scripts sourced from 2.4 million products. Human annotations are then conducted to provide gold labels for a sampled subset, forming an evaluation benchmark. Extensive experiments reveal that current (L)LMs face significant challenges with EcomScript tasks, even after fine-tuning, while injecting product purchase intentions improves their performance.

Rosa Santero-Sánchez, Belén Castro Núñez

Despite the massive incorporation of women into the labor market, equal pay for equal work remains a challenge. This article analyzes the influence of gender diversity in management positions on the gender wage gap (GWG) throughout the entire pay scale in Spain. The results show the existence of a GWG, particularly for wages below the average; though it decreases when female participation in management is higher. This is in line with the reduction of information asymmetry problems considered in statistical discrimination theories, which explain the barriers to promotion associated with dynamics at entry-level and low-qualified positions. JEL CLASSIFICATION: C31, J31, J71, M14

Anisur Rehman, Aftab Ara, Harman Preet Singh

Abstract Saudi Arabia ranks amongst those nations with the highest amount of fossil carbon dioxide emissions. Due to the rising environmental and economic sustainability challenges in Saudi Arabia, investigating environmental sustainability in the nation has become imperative. Similar to business organisations, universities in Saudi Arabia are also intensifying their initiatives to enhance sustainability, contending with the nation’s established dependence on increasing greenhouse gas emissions, high energy and fossil fuels. There is a gap in research on the green human resource management (GHRM) practices of these universities aimed at promoting sustainability with the involvement of their faculty members. To address this gap, this study employs ability–motivation–opportunity (AMO) theory and change management theory to examine the effect of GHRM practices on the organisational citizenship behaviour towards environment (OCBE) and environmental performance (EP) of universities in Saudi Arabia. This study also investigates the moderating effect of technological competence and resistance to change on the association of GHRM practices with OCBE and EP. Results demonstrate that all three GHRM practices (green training and development, green employee involvement and green performance appraisal) have a positive and significant impact on OCBE, which subsequently influences EP. The mediating effect of OCBE and the moderating effects of technological competence and resistance to change were also confirmed. This study advances the existing research on the influence of GHRM practices on environmental outcomes through the lens of the AMO framework and change management theory in the university context of Saudi Arabia. This pioneering study also explores the role of technological competence and resistance to change as moderators in the above relationship. In addition to theoretical implications, this study offers novel insights for policymakers in higher educational institutions.

Tayeb AMARA

تتجلى أهمية الدراسة في إبراز وظيفة التكوين في كونها أساس التفوق و تأكيد القدرة على مسايرة التحديات بل و تجاوزها في مخبر  صناعة الموارد البشرية في المصارف العمومية الجزائرية. من أجل ذلك، كان  لزاما على وظيفة التكوين ضمان النتائج الآنية و المستقبلية المتبناة في السياسات العامة. و لقد بينت عملية استطلاع آراء عينة مكونة من ( 107) إطار مركزي و جهوي متخصص في تسيير الموارد البشرية، و من مخرجات البرنامج الإحصائي (SPSS.V.21)، أن هناك علاقة ذات أثر ذو دلالة إحصائية بين  وظيفة التكوين و بين المردودية المنتظرة من الموارد البشرية،  دونما وجود فروق ذات دلالة إحصائية حول مدى التزام مديري مراكز التكوين بتطبيق المناهج التي تعزز مكانة وظيفة التكوين، تعزى لعوامل المؤهل المهني، المؤهل العلمي، المنصب الحالي، الأقدمية.

Mauro Oddo Nogueira

Objetivo do Estudo: Evidenciar o papel central das micro e pequenas empresas (MPEs) para a superação da perversa desigualdade socioeconômica brasileira, destacando a necessidade de que recebam tratamento prioritário nas políticas públicas. Principais resultados: Ressaltando a desconhecida e negligenciada realidade das MPEs, demonstra que a produtividade é o principal desafio, pois a maioria das MPEs apresenta baixíssimos níveis de produtividade, operando em um ambiente de informalidade/semiformalidade, agravando o dilema produtivo do país, limitando seu potencial de crescimento e as possibilidades de superação da desigualdade. Mostra, ainda, que, embora as MPEs formais e informais, representem a parcela mais significativa da economia em termos de PIB e ocupações, são tratadas marginalmente, não recebendo do Estado, da academia ou da mídia, atenção compatível com essa importância. Relevância/originalidade: Traz uma crítica original à visão que reduz o empreendedorismo à criação de novos negócios, tratando-o como panaceia para os problemas nacionais. Propõe, em reverso, a requalificação dos empreendedores já existentes e o apoio a inovações que aumentem o conteúdo técnico dos postos de trabalho (modernização de processos de produção e gestão), resultando no aumento da produtividade e da competitividade das MPEs. Contribuições sociais/para a gestão: Sugere a formulação de políticas públicas que atribuam centralidade às MPEs o que resultaria em um processo de desenvolvimento inclusivo e sustentável. Além disso, enfatiza a premência de mais estudos sobre MPE e informalidade, para se compreender adequadamente a realidade desse segmento vital da economia brasileira

Ana-Maria Coatu, Felix-Angel Popescu, Laurențiu Petrila

This study explores how socio-economic factors affect the effectiveness of public accountability frameworks in EU member states, with Romania as a case study. Using data from the World Bank, Eurobarometer, and cross-country comparisons, it identifies five key determinants: income inequality, education, healthcare access, political participation, and economic stability. Grounded in institutional theory, the research shows that inclusive institutions and lower disparities lead to stronger accountability, while weaker frameworks often reinforce inequality and corruption. For Romania, the study recommends boosting transparency, enforcing anti-corruption measures, improving rural-urban equity, and enhancing civic education to strengthen the link between citizens and institutions.

Soheila Khajoui, Saeid Dehyadegari, Sayyed Abdolmajid Jalaee

This study aims at predicting the impact of e-commerce indicators on international trade of the selected OECD countries and Iran, by using the artificial intelligence approach and P-VAR. According to the nature of export, import, GDP, and ICT functions, and the characteristics of nonlinearity, this analysis is performed by using the MPL neural network. The export, import, GDP, and ICT findings were examined with 99 percent accuracy. Using the P-VAR model in the Eviews software, the initial database and predicted data were applied to estimate the impact of e-commerce on international trade. The findings from analyzing the data show that there is a bilateral correlation between e-commerce which means that ICT and international trade affect each other and the Goodness of fit of the studied model is confirmed.

Yudina Nurhaliza, Dwi Novita Sari, Maisyahrani Maisyahrani et al.

This research aims to analyze whether mudharabah and murabaha financing have an impact on maximizing real sector growth in Indonesia, Malaysia and Brunei Darussalam. The method used in this research is quantitative with a descriptive-associative approach. The data used in this research comes from secondary data and time series data obtained from the Financial Services Authority (FSA) Islamic banking statistics report, Indonesian economic and financial statistics report, Kuwait Finance House annual report, Malaysian economy in Figures 2022 report, report annual Islamic bank Brunei Darussalam, and the International Monetary Fund (IMF) report. This research uses panel data regression analysis. The research results show that partial mudharabah and murabaha financing does not impact real sector growth. Simultaneously, mudharabah and murabaha financing significantly impact real sector growth. This research provides a new contribution to scientific development, stating that mudharabah and murabaha financing do not impact real sector growth. Practically, this research proves that there must be improvements related to mudharabah and murabaha financing so that they can have an impact on real sector growth.

Inayatul Sabilla Azahro, Diah Hari Suryaningrum

Riset ini mencoba untuk menguji korelasi antara kesulitan keuangan perusahaan dan manipulasi laporan keuangan dengan menggunakan teknik manajemen laba. Selain itu, penelitian ini juga bermaksud untuk mengungkap hubungan antara kesulitan keuangan dan manajemen laba. Altman Z-score digunakan untuk mengevaluasi kondisi kesehatan keuangan perusahaan, sedangkan Beneish M-score dan model Jones yang dimodifikasi digunakan untuk memastikan adanya praktik manajemen laba. Laporan keuangan untuk periode 2018-2022 dari 567 perusahaan di Bursa Efek Indonesia dikumpulkan sebagai sampel berdasarkan mekanisme purposive sampling. Uji chi-square Pearson, uji Cramer's V, dan analisis korespondensi diterapkan guna menguji hipotesis yang ada. Buah dari riset ini membawa hasil yakni adanya kaitan antara financial distress dan praktik manajemen laba. Selain itu, perusahaan yang berada di zona abu-abu dan zona sehat lebih cenderung melakukan manipulasi laba. Penelitian ini mengimplikasikan bahwa financial distress dan manajemen laba dapat diukur dan dideteksi oleh siapa saja sehingga seluruh pengguna laporan keuangan tidak akan dirugikan ketika terjadi kecurangan.

Jie Huang, Yifan Gao, Zheng Li et al.

We propose and study Complementary Concept Generation (CCGen): given a concept of interest, e.g., "Digital Cameras", generating a list of complementary concepts, e.g., 1) Camera Lenses 2) Batteries 3) Camera Cases 4) Memory Cards 5) Battery Chargers. CCGen is beneficial for various applications like query suggestion and item recommendation, especially in the e-commerce domain. To solve CCGen, we propose to train language models to generate ranked lists of concepts with a two-step training strategy. We also teach the models to generate explanations by incorporating explanations distilled from large teacher models. Extensive experiments and analysis demonstrate that our model can generate high-quality concepts complementary to the input concept while producing explanations to justify the predictions.

Zhao-Yang Liu, Liucheng Sun, Chenwei Weng et al.

Bundle recommendation aims to provide a bundle of items to satisfy the user preference on e-commerce platform. Existing successful solutions are based on the contrastive graph learning paradigm where graph neural networks (GNNs) are employed to learn representations from user-level and bundle-level graph views with a contrastive learning module to enhance the cooperative association between different views. Nevertheless, they ignore the uncertainty issue which has a significant impact in real bundle recommendation scenarios due to the lack of discriminative information caused by highly sparsity or diversity. We further suggest that their instancewise contrastive learning fails to distinguish the semantically similar negatives (i.e., sampling bias issue), resulting in performance degradation. In this paper, we propose a novel Gaussian Graph with Prototypical Contrastive Learning (GPCL) framework to overcome these challenges. In particular, GPCL embeds each user/bundle/item as a Gaussian distribution rather than a fixed vector. We further design a prototypical contrastive learning module to capture the contextual information and mitigate the sampling bias issue. Extensive experiments demonstrate that benefiting from the proposed components, we achieve new state-of-the-art performance compared to previous methods on several public datasets. Moreover, GPCL has been deployed on real-world e-commerce platform and achieved substantial improvements.

Hai Zhu, Yuankai Guo, Ronggang Dou et al.

Relevance module plays a fundamental role in e-commerce search as they are responsible for selecting relevant products from thousands of items based on user queries, thereby enhancing users experience and efficiency. The traditional approach models the relevance based product titles and queries, but the information in titles alone maybe insufficient to describe the products completely. A more general optimization approach is to further leverage product image information. In recent years, vision-language pre-training models have achieved impressive results in many scenarios, which leverage contrastive learning to map both textual and visual features into a joint embedding space. In e-commerce, a common practice is to fine-tune on the pre-trained model based on e-commerce data. However, the performance is sub-optimal because the vision-language pre-training models lack of alignment specifically designed for queries. In this paper, we propose a method called Query-LIFE (Query-aware Language Image Fusion Embedding) to address these challenges. Query-LIFE utilizes a query-based multimodal fusion to effectively incorporate the image and title based on the product types. Additionally, it employs query-aware modal alignment to enhance the accuracy of the comprehensive representation of products. Furthermore, we design GenFilt, which utilizes the generation capability of large models to filter out false negative samples and further improve the overall performance of the contrastive learning task in the model. Experiments have demonstrated that Query-LIFE outperforms existing baselines. We have conducted ablation studies and human evaluations to validate the effectiveness of each module within Query-LIFE. Moreover, Query-LIFE has been deployed on Miravia Search, resulting in improved both relevance and conversion efficiency.