Overview
The “Gen AI for E-commerce” workshop explores the role of Generative Artificial Intelligence in transforming e-commerce through enhanced user experience and operational efficiency. E-commerce companies grapple with multiple challenges such as lack of quality content for products, subpar user experience, sparse datasets etc. Gen AI offers significant potential to address these complexities. Yet, deploying these technologies at scale presents challenges such as hallucination in data, excessive costs, increased latency response, and limited generalization in sparse data environments. This workshop will bring together experts from academia and industry to discuss these challenges and opportunities, aiming to showcase case studies, breakthroughs, and insights into practical implementations of Gen AI in e-commerce.
Call for papers
We will welcome papers that leverage Generative Artificial Intelligence (Gen AI) in e-commerce. Detailed topics are mentioned in CFP. Papers can be submitted at Easychair.
Information for the day of the workshop
Workshop at RecSys2025
- Paper submission deadline: August 10, 2025
- Paper acceptance notification: August 25, 2025
- Camera-ready manuscript deadline: September 4, 2025
- Workshop: September 22, 2025
Schedule
We have a half-day program on September 22 at Prague, Czech Republic.
| Time (CET) | Agenda |
|---|---|
| 1:30-1:40pm | Registration and Welcome |
| 1:40-1:55pm | Paper Presentation: Mind the Gap: Bridging Behavioral Silos with LLMs in Multi-Vertical Recommendations |
| 1:55-2:10pm | Paper Presentation: Image-Seeking Intent Prediction for Cross-Device Product Search |
| 2:10-2:25pm | Paper Presentation: Using item recommendations and LLMs in marketing email titles |
| 2:25-2:40pm | Paper Presentation: Enhancing Related Searches Recommendation System by Leveraging LLM Approaches |
| 2:40-2:55pm | Paper Presentation: Knowledge-Augmented Relation Learning for Complementary Recommendation with Large Language Models |
| 2:55-3:10pm | Paper Presentation: TOD-ProcBench: Benchmarking Operating Procedure Compliance in Task-Oriented Dialogues |
| 3:10-3:25pm | Paper Presentation: CRMAgent: A Multi-Agent LLM System for E-Commerce CRM Message Template Generation |
| 3:30-4:00pm | Coffee Break |
| 4:00-4:40pm | Keynote by Chen Luo : From Search to Conversational Shopping with Generative AI |
| 4:40-5:30pm | Poster Session |
Keynote Speakers
Chen Luo
Amazon
From Search to Conversational Shopping with Generative AI
Abstract: The rapid growth of online shopping platforms such as Amazon has brought services to billions of people worldwide. With global retail sales surpassing $6 trillion in 2024 and is trending increase in 2025, customer expectations for personalized and seamless shopping experiences have never been higher. In this talk, I will share our multi years journey of reimagining online shopping at Amazon—transforming it from a product search problem into a conversational shopping experience powered by generative models. Specifically, I will cover: (1) How the product search engine traditionally worked. (2) How we began adopting conversational shopping and use language models for this new paradigm. (3) How we are building shopping agents that empower customers with richer, more personalized experiences.
Bio
Bio: Chen is Sr. Applied Scientist at Amazon Search (previously known as A9). He obtained his Ph.D. from Rice University, working with Anshumali Shrivastava. Before Rice, he was a master student in Key Laboratory of Symbolic Computation and Knowledge Engineering, Jilin University. He recieved his B.S degree from the Department of Computer Science, Jilin University.
Accepted Papers
- Mind the Gap: Bridging Behavioral Silos with LLMs in Multi-Vertical Recommendations
Nimesh Sinha, Raghav Saboo, Martin Wang, Sudeep DasAbstractAbstract: In multi-vertical e-commerce platforms like DoorDash, relatively newer product verticals such as grocery and retail present a significant opportunity for personalization innovation. A key challenge lies in solving the "cold start" problem for users. This paper introduces a novel framework for enhancing recommendation quality by transferring knowledge from data-rich verticals (e.g., restaurants at DoorDash) to data-sparse ones. We leverage Large Language Models (LLMs) to perform generative inference, synthesizing sparse, high-dimensional features that encapsulate user preferences across different verticals.PDF Code - Image-Seeking Intent Prediction for Cross-Device Product Search
Mariya Hendriksen, Svitlana Vakulenko, Jordan Massiah, Gabriella Kazai, Emine YilmazAbstractAbstract: Abstract to be provided by authors.PDF Code - Using item recommendations and LLMs in marketing email titles
Deddy Jobson, Muktti Shukla, Phuong Dinh, Julio Christian Young, Nick Pittoni, Nina Chen, Ryan GinstromAbstractAbstract: Abstract to be provided by authors.PDF Code - Knowledge-Augmented Relation Learning for Complementary Recommendation with Large Language Models
Chihiro Yamasaki, Kai Sugahara, Kazushi OkamotoAbstractAbstract: Abstract to be provided by authors.PDF Code - TOD-ProcBench: Benchmarking Complex Instruction-Following in Task-Oriented Dialogues
Nanyun Peng, Narayanan Sadagopan, Zhou YuAbstractAbstract: Abstract to be provided by authors.PDF Code - Mind the Gap: Linguistic Divergence and Adaptation Strategies in Human-LLM Assistant vs. Human-Human Interactions
Fulei Zhang, Zhou YuAbstractAbstract: Abstract to be provided by authors.PDF Code - CRMAgent: A Multi-Agent LLM System for E-Commerce CRM Message Template Generation
Yinzhu Quan, Xinrui Li, Ying ChenAbstractAbstract: In e-commerce private-domain channels such as instant messaging and e-mail, merchants engage customers directly as part of their Customer Relationship Management (CRM) programmes to drive retention and conversion. While a few top performers excel at crafting outbound messages, most merchants struggle to write per- suasive copy because they lack both expertise and scalable tools. We introduce CRMAgent, a multi-agent system built on large language models (LLMs) that generates high-quality message templates and actionable writing guidance through three complementary modes. First, group-based learning enables the agent to learn from a merchant’s own top-performing messages within the same audience segment and rewrite low-performing ones. Second, retrieval-and- adaptation fetches templates that share the same audience segment and exhibit high similarity in voucher type and product category, learns their successful patterns, and adapts them to the current cam-paign. Third, a rule-based fallback provides a lightweight zero-shot rewrite when no suitable references are available. Extensive experi- ments show that CRMAgent consistently outperforms merchants’ original templates, delivering significant gains in both audience- match and marketing-effectiveness metrics.PDF Code - Enhancing Related Searches Recommendation system by leveraging LLM Approaches
Hung Nguyen, Jayanth Yetukuri, Phuong Ha Nguyen, Lizzie Liang, Ishita Khan, Zhe WuAbstractAbstract: Abstract to be provided by authors.PDF Code
Organizers
Mansi Mane
Walmart Global Tech
Bio: Mansi Mane is Staff Data Scientist at Search and Recommendation team in Walamrt Labs. She completed her Masters from Carnegie Mellon University in 2018. She currently focuses on research and development of machine learning algorithms for recommendations, search, marketing as well as content generation. Mansi was previously Applied Scientist at AWS where she lead efforts for training of billion scale large language models from scratch. Her research interests include machine learning, multimodal LLMs pretraining, fine-tuning as well as in-context learning.
Djordje Gligorijevic
eBay
Bio: Djordje Gligorijevic is applied sciences manager at eBay, leading allocation and pricing team in eBay’s sponsored search program. Prior to eBay Djordje worked as Research Scientist in Yahoo Research. He received the Ph.D. degree from Temple University, Philadelphia, PA, in 2018. His research interests include Machine Learning, Extreme Multi-Label Classification, NLP, LLMs, and the Integration of Qualitative Knowledge into predictive models with applications in domains of Computational Healthcare, Computational Advertising, Search, Ranking, and Recommendation Systems. Djordje has published at international conferences such as AAAI, KDD, TheWebConf, SDM, CIKM, SIGIR, as well as in international journals like Data Mining and Knowledge Discovery, BigData journal where he serves as associate editor, Methods and Nature’s Scientific Reports.
Dingxian Wang
Upwork
Bio: Dingxian is a Applied Science leader with around 10 years industry experience at the intersection of machine learning, software engineering, applied science, and product development. He is passionate about applying skills to solving real-world problems, especially in the field of technology and data science. He is currently leading a team focus on the ranking, personalization and recommendation in the search area. Throughout the career, Dingxian has been involved in a wide range of areas, including search engine, query understanding, recall system, ranking system, recommender system, marketing science, personalization, information extraction, knowledge graph etc. With massive proven track records of delivering great business results, and drove hundreds of millions of dollars in GMV and revenue growth. Dingxian has received many top honors and awards ranging from top conference, journals, patents to top research projects as well as internal competition awards. Including 20+ papers on top conference and journals (one best paper candidate of CIKM 2021), 9 US patents, over 1500 citations, ICT Research Project of the Year 2021 of ACS (Australian Computer Society) and eBay Leaders’ Choice Award.
Topojoy Biswas
Walmart Global Tech
Bio: Topojoy Biswas is Distinguished Data Scientist at Walmart Labs. At Walmart he leads efforts related to W+ membership models and creative generation projects. Prior to Walmart he worked as Principal Engineer at Yahoo Research where he worked on information extraction on text and videos in Yahoo Knowledge Graph which powers search and information organization in products in Yahoo, like Finance, Sports, entity search and browse. Before Yahoo Knowledge graphs, he worked for Yahoo shopping on attribute extraction and classification of shopping feeds into large taxonomies of products. Topojoy has published in multiple international conferences such as ICIP, ACM Multimedia etc and has spoken on applied machine learning topics in MLConf, KGC etc.
Evren Korpeoglu
Walmart Global Tech
Bio: Evren Korpeoglu is a Director of Data Science at Personalization and Recommendations team in Walmart Global Tech. At Walmart he leads efforts related to whole page optimization, item recommendations as well as using Generative AI based models for recommendations. He completed his Ph.D. from Bikent University. He has published at international conferences like NeurIPS, ICML, SIGKDD.
Yongfeng Zhang
Rutgers University
Bio: Yongfeng Zhang is an Assistant Professor in the Department of Computer Science at Rutgers University (The State University of New Jersey). His research interest is in Information Retrieval, Economics of Data Science, Explainable AI, Fairness in AI and AI Ethics. In the previous he was a postdoc advised by Prof. W. Bruce Croft in the Center for Intelligent Information Retrieval (CIIR) at UMass Amherst, and did his PhD and BE in Computer Science at Tsinghua University, with a BS in Economics at Peking Univeristy. He is a Siebel Scholar of the class 2015. Together with coauthors, he has been consistently working on explainable search and recommendation models, fairness-aware machine learning, conversational search and recommendation, neural logic/symbolic reasoning, economic machine learning , efficient/robust machine learning, knowledge graph embedding, legal/medical retrieval, as well as causal/counterfactual models for information retrieval. His recent research on causality in search and recommendation include causal collaborative filtering, causal explainable recommendation, counterfactual debiasing models, and causal models for mitigating feedback loops in IR systems.
Marios Savvides
CMU, UltronAI
Bio: Professor Marios Savvides is the Bossa Nova Robotics Professor of Artificial Intelligence at Carnegie Mellon University and is also the Founder and Director of the Biometrics Center at Carnegie Mellon University and a Full Tenured Professor in the Electrical and Computer Engineering Department. He received his Bachelor of Engineering in Microelectronics Systems Engineering from University of Manchester Institute of Science and Technology in 1997 in the United Kingdom, his Master of Science in Robotics from the Robotics Institute in 2000 and his PhD from the Electrical and Computer Engineering department at CMU in 2004. He has authored and co-authored over 250 journal and conference publications, including 22 book chapters and served as the area editor of the Springer’s Encyclopedia of Biometrics. Some of his notable accomplishments include developing a 40ft stand-off distance iris recognition system, robust face detection even in presence of extreme occlusions, a fully autonomous AI inventory robotic image analytics system for detecting out-of-stocks that he and his team scaled to 550 walmart retail stores. His latest research is in large foundation vision models for zero shot enrollment for robust object recognition which has spun out as the enterprise software company UltronAI, Inc. His work was presented at the World Economic Forum (WEF) in Davos, Switzerland in January 2018 and his research has been featured in over 100 news media articles (CNN, CBS 60 minutes, Scientific American, Popular Mechanics etc). He is the recipient of CMU’s 2009 Carnegie Institute of Technology (CIT) Outstanding Research Award, the Gold Award in the 2015 Edison Awards in Applied Technologies for his biometrics work, 2018 Global Pittsburgh Immigrant Entrepreneur Award in Technological Innovation, the 2020 Artificial Intelligence Excellence Award in “Theory of Mind”, the Gold Award in 2020 Edison Awards for Retail Innovations on Autonomous Data Capture and Analysis of On-Shelf Inventory, the “Stevens J. Fenves Award for Systems Research”, the “2020 Outstanding Contributor to AI” award from the US Secretary of the Army Mr. Ryan McCarthy, named the “Inventor of Year” by the Pittsburgh Intellectual Property Law Association (PIPLA), 2022.
Julian McAuley
UC San Diego
Bio: McAuley is a professor in CSE. His research focuses on the linguistic, temporal, and social dimensions of opinions and behavior in social networks and other online communities. He has harnessed data science tools to increase understanding the facets of people’s opinions, the processes that lead people to acquire taste for gourmet foods and beers, and even the visual dimensions of how they make fashion choices. He has gained academic, industry and media attention for his work analyzing massive volumes of user data from online social communities including Amazon, Yelp, Facebook and BeerAdvocate. His work includes using artificial intelligence in fashion choice, and data science in developing models that generate step-charts for the globally popular videogame, Dance Dance Revolution.
Program Committee
- Sample PC Member (Sample University)