Reza Yousefi Maragheh

Reza Yousefi Maragheh

Ph.D. Data Science at Walmart Labs

About Me

I am Reza Yousefi Maragheh, a Staff Data Scientist working at Walmart Labs’-Personalization Team. I got my Ph.D from the University of Illinois Urbana-Champaign (UIUC)-Grainger College of Engineering. My Ph.D. thesis was on “Choice Modeling and Recommendation Optimization in Presence of Context Effects” (check it out here) supervised by Professor Xin Chen. My Research interests are Recommender Systems, Information Retrieval, NLP for RecSys, Customer Intent Inference, and Optimization. In the past two years, I was specifically involved in exploring/implementing large-scale language agent pipelines in classic ecormmerce tasks.

News:

  • May 2026: Our tutorial on Production-Grade Agentic Recommender Systems has been accepted to SIGIR 2026 in Melbourne, Australia (Presentation this July). The GitHub page is now live — it also hosts materials from our RecSys 2025 tutorial: https://agenticrecsys.github.io/. Instructors: Reza Yousefi Maragheh (Walmart Labs), Yashar Deljoo (Politecnico di Bari), Ben Coleman (DeepMind), Jason Cho (pretaskAI), and Chi Wang (AG2AI / DeepMind).
  • Dec 2025: We tested Multi-Agentic Evaluation Framework for reducing output variance in production. Results published in NeurIPS 2025 Workshop on Multi-Turn Interactions in Large Language Models. link. Authors: Reza Yousefi Maragheh et al.
  • Dec 2025: We productionalized a Text Generation framework with Implicit Feedback. Result published in IEEE BIGDATA 25 and NeurIPS25 workshop on Bridging Language, Agent, and World Models for Reasoning and Planning. Authors: Reza Yousefi Maragheh et al.
  • Sep 2025: Content for Agentic RecSys Tutorial is available here: https://agenticrecsys.github.io/
  • Sep 2025: We will have a tutorial on “Multi Agentic Recommender Systems: Foundations, Design Patterns, and E-Commerce Applications” with Yashar Deljoo (University of Bari, Italy), Jason Cho (Walmart Labs), Derek Cheng (Google Deepmind), and Chi Wang (Google Deepmind) in RecSys Conference this September in Prague, Czech Republic. link
  • Sep 2025: I am invited to Give a lecture on Modern Topics on Recommender Systems in RecSys Summer School in Vienna, Austria (Sep 17th). link
  • July 2025: A Review of our ARAG work is published by Elvis Saravia, Quantum Zeitgeist, and Top Information Retrieval Papers of the Week.
  • July 2025: Our paper with the title ARAG: Agentic Retrieval Augmented Generation for Personalized Recommendation will be presented in SIGIR, Link,Authors: Reza Yousefi Maragheh et al.
  • July 2025: Our paper with the title CAL-RAG: Retrieval-Augmented Multi-Agent Generation for Content-Aware Layout Design will be presented in SIGIR, Link Authors: Najmeh Forouzandemehr and Reza Yousefi Maragheh et al.
  • July 2025: Our paper with the title CARTS: Collaborative Agents for Recommendation Textual Summarization will be presented in ICML, Link Jiao Chen and Reza Yousefi Maragheh et al.. The work is based on a recent user facing product launch.
  • April 2025: We are invited to run an Q&A session at Google Next Conference on Agentic AI applications in eCommerce domain, link
  • Aug 2024: Our innovation in optimized training algorithm for mnl based models, titled ‘Systems and methods for hybrid optimization training of multinomial logit models’, has been filed with the United States Patent and Trademark Office (USPTO). Link-Authors: Reza Yousefi Maragheh et al.
  • Aug 2024: Our innovation in personalized recommendation systems, titled ‘System and method for recommending items based on enhanced user representations,’ has been filed with the United States Patent and Trademark Office (USPTO). Link-Authors: Aysenur Inan and Reza Yousefi Maragheh et al.
  • Dec 2023: Our Paper “Seller-side Outcome Fairness in Online Marketplaces” is presented at NIPS 2023, Link-Authors:Zikun Ye and Reza Yousefi Maragheh et al.
  • Dec 2023: “LLM-TAKE: Theme-Aware Keyword Extraction Using Large Language Models” is presented at IEEE Big Data 2023, Link. Authors: Reza Yousefi Maragheh et al.
  • Nov 2023: Our Paper “GNN-GMVO: Graph Neural Networks for Optimizing Gross Merchandise Value in Similar Item Recommendation” is presented at ICDM 2023, Link. Authors: Ramin Giahi and Reza Yousefi Maragheh et al.
  • June 2023: Our paper, LLM-Based Aspect Augmentations for Recommendation Systems, is mentioned by Sumit Gupta, VP at Google: link
  • June 2023: Our paper, LLM-Based Aspect Augmentations for Recommendation Systems, is mentioned by Dylan Patel, Technology Reviewer, and Consultant: link
  • June 2023: Our Paper “LLM-Based Aspect Augmentations for Recommendation Systems” is presented at ICML 2023, Link. Authors: Reza Yousefi Maragheh et al.