Department of Computer Science

The Amazon Research Awards program has provided $100,000 in funding to a groundbreaking project led by Ping Wang, an assistant professor in Stevens’ Department of Computer Science. The project, titled "Leveraging Large Language Models for Reasoning Augmented Searching on Domain-Specific NoSQL Database," aims to advance natural language querying (NLQ) techniques for NoSQL databases.

The research focuses on using modern deep learning and semantic parsing techniques to translate natural language questions into structured query language (SQL) queries. Despite the prevalence of efficient NoSQL databases, there has been limited development of NLQ tools for these systems. Wang's team will conduct a comprehensive analysis of NoSQL databases and integrate recent advances in multi-hop reasoning on knowledge graphs and large language models (LLMs) to propose a reasoning-augmented searching framework for ElasticSearch.

This innovative project promises to enhance NLQ techniques on NoSQL databases, offering significant benefits for various real-world applications. The funding from Amazon Research will support the team's efforts to bridge the gap in NLQ tool development and exploit the potential of NoSQL databases.

Zhaozhuo Xu, an assistant professor in the Department of Computer Science, is among the select few early career faculty members worldwide to be selected as part of the New Faculty Highlights program at the 2025 Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence.

The New Faculty Highlights program highlights AI researchers who have just begun careers as new faculty members or the equivalent in industry. New Faculty Highlight talks will be allotted 30 minutes each, providing the candidates a stage to broadly survey their research to date. Selections are based on significance and impact of research, vision and leadership potential, quality of publications and more.

Xu’s talk, titled “Compression-Aware Computing for Scalable and Sustainable AI,” explores the challenge of customizing large-scale AI models, particularly generative AI, on cost-effective devices with limited memory and energy resources. The talk introduces compression-aware computing, a framework enabling AI models to recognize and adapt to their compressed states while preserving performance. Compression-aware computing integrates compression techniques like sparsification, quantization and low-rank decomposition to enhance the efficiency and accuracy of AI models, broadening these models’ accessibility across diverse devices.

The talk will also highlight his research’s potential regarding Alzheimer’s disease by facilitating the analysis of large single-cell transcriptomics datasets for gene-gene interaction discovery.

The conference will take place Feb. 25-March 4 in Philadelphia.

David Naumann, professor and chair of the Department of Computer Science, has received a $588,353 grant from the National Science Foundation (NSF) for his project titled “SaTC: CORE: Small: Auto-Active Hyperproperty Verification for Security.”

In this project, verification refers to proof of software correctness with respect to security and privacy requirements that are precisely specified using mathematical models of environments that include active adversaries. Auto-active verification leverages automated theorem provers to enable proving to be integrated into software development tools and processes. Auto-active verification is increasingly used in industry, but so far it has been limited to functional requirements that can be specified in terms of individual executions. Information-flow requirements must be specified in terms of relationships between multiple executions, called relational properties or hyperproperties. Naumann’s project aims to develop theory and techniques to enable auto-active verification of relational properties.

The project will develop logics and methods for reasoning about refinement, equivalence and hyperproperties. The theoretical advances will be realized in prototype tools used to evaluate usability and effectiveness. Naumann believes this research has the potential to transform computing practice by supporting accountability of system designers and builders through mathematically precise specifications and machine-checked proofs of security and privacy. This should improve software security and quality assurance. Equivalence verification may lessen the risk that code improvements and big fixes will introduce new vulnerabilities or adversely affect system functionality.

Tian Han, assistant professor of computer science at Stevens, recently received a five-year, $599,729 grant from the National Science Foundation (NSF) CAREER Award program to explore generative AI.

The Faculty Early Career Development (CAREER) Program is a foundation-wide activity that offers the National Science Foundation's most prestigious awards in support of early-career faculty who have the potential to serve as academic role models in research and education.

Han aims to overcome three of the most significant hurdles regarding generative AI, which leverages advanced learning models to generate content such as articulate text and realistic images and videos.

“First, we will work to extract useful information from high-dimensional data such as images or text to learn structured knowledge for smart generative AI models,” said Han. “Then we will create controllable models that allow users to adjust both the models and the results, such as tweaking the intensity of a smile or adding a hat in a photo. Finally, we will design general, flexible frameworks that can simultaneously process data from different domains, such as text and images, to deliver consistent content.”

The grant also includes funding to train undergraduate, graduate and Ph.D. students. Research topics and findings will be used to further strengthen the AI curriculum at Stevens to continue to help lead the future of generative AI.