Gediminas Adomavicius’s research while affiliated with University of Minnesota, Duluth and other places

Large Language Models (LLMs) are being applied in a wide array of settings, well beyond the typical language-oriented use cases. In particular, LLMs are increasingly used as a plug-and-play method for fitting data and generating predictions. Prior work has shown that LLMs, via in-context learning or supervised fine-tuning, can perform competitively with many tabular supervised learning techniques in terms of predictive performance. However, we identify a critical vulnerability of using LLMs for data fitting -- making changes to data representation that are completely irrelevant to the underlying learning task can drastically alter LLMs' predictions on the same data. For example, simply changing variable names can sway the size of prediction error by as much as 82% in certain settings. Such prediction sensitivity with respect to task-irrelevant variations manifests under both in-context learning and supervised fine-tuning, for both close-weight and open-weight general-purpose LLMs. Moreover, by examining the attention scores of an open-weight LLM, we discover a non-uniform attention pattern: training examples and variable names/values which happen to occupy certain positions in the prompt receive more attention when output tokens are generated, even though different positions are expected to receive roughly the same attention. This partially explains the sensitivity in the presence of task-irrelevant variations. We also consider a state-of-the-art tabular foundation model (TabPFN) trained specifically for data fitting. Despite being explicitly designed to achieve prediction robustness, TabPFN is still not immune to task-irrelevant variations. Overall, despite LLMs' impressive predictive capabilities, currently they lack even the basic level of robustness to be used as a principled data-fitting tool.

As electric vehicles (EVs) become more widespread, cities face the growing challenge of managing charging demand without overloading the grid. This study presents a novel information systems (IS) solution that supports smart and sustainable EV integration. The authors develop a capacity-based pricing model that adjusts in real time based on charging rates and grid capacity. Unlike many existing approaches, it avoids “avalanche effects” where synchronized charging behavior creates new demand peaks. The presented solution is also computationally efficient, making it practical for real-world use. Evaluated through simulations based on realistic urban scenarios, the model reduces demand volatility, aligns EV charging with renewable energy availability, and maintains overall charging costs for users. This work offers policy makers and energy providers a concrete tool to balance environmental goals with energy system reliability. For urban mobility planners, it provides a scalable, adaptive method to support the transition to cleaner urban mobility.

Multistakeholder recommender systems are those that account for the impacts and preferences of multiple groups of individuals, not just the end users receiving recommendations. Due to their complexity, evaluating these systems cannot be restricted to the overall utility of a single stakeholder, as is often the case of more mainstream recommender system applications. In this article, we focus our discussion on the intricacies of the evaluation of multistakeholder recommender systems. We bring attention to the different aspects involved in the evaluation of multistakeholder recommender systems - from the range of stakeholders involved (including but not limited to producers and consumers) to the values and specific goals of each relevant stakeholder. Additionally, we discuss how to move from theoretical principles to practical implementation, providing specific use case examples. Finally, we outline open research directions for the RecSys community to explore. We aim to provide guidance to researchers and practitioners about how to think about these complex and domain-dependent issues of evaluation in the course of designing, developing, and researching applications with multistakeholder aspects.

Multistakeholder recommender systems are defined as those that account for ``the preferences of multiple parties when generating recommendations, especially when these parties are on different sides of the recommendation interaction.'' Due to their complexity, evaluating these systems cannot be restricted to the overall utility of a single stakeholder, as is often the case of more mainstream recommender system applications. In this chapter, we focus our discussion on the intricacies involved in understanding what is the ``right'' construct required to ensure the proper evaluation of multistakeholder recommender systems. We bring attention to the different aspects involved in the evaluation of multistakeholder recommender systems---from the range of stakeholders involved (beyond producers and consumers) to the values and specific goals of each relevant stakeholder. Additionally, we discuss how to move from theoretical evaluation to practical implementation, providing specific use case examples. Finally, we outline open research directions for the RecSys community to explore. Our aim in this section is to provide guidance to researchers and practitioners about how to think about these complex and domain-dependent issues in the course of designing, developing, and researching applications with multistakeholder aspects. This chapter is part of the following Dagstuhl report: https://www.dagstuhl.de/de/seminars/seminar-calendar/seminar-details/24211

This archive is distributed in association with the INFORMS Journal on Computing under the MIT License. The purpose of this repository is to share the data and scripts reported in the paper Efficient and Flexible Long-Tail Recommendation Using Cosing Patterns by Y. Wang, J. Wu, Z. Wu, and G. Adomavicius.

With the increasing use of recommender systems in various application domains, many algorithms have been proposed for improving the accuracy of recommendations. Among various dimensions of recommender systems performance, long-tail (niche) recommendation performance remains an important challenge in large part because of the popularity bias of many existing recommendation techniques. In this study, we propose CORE, a cosine pattern–based technique, for effective long-tail recommendation. Comprehensive experimental results compare the proposed approach with a wide variety of classic, widely used recommendation algorithms and demonstrate its practical benefits in accuracy, flexibility, and scalability in addition to the superior long-tail recommendation performance. ¹ History: Accepted by Ramaswamy Ramesh, Area Editor for Data Science & Machine Learning. Funding: This work was supported by the National Natural Science Foundation of China [Grants 72031001, 72072091, 72242101]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0194 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0194 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .