The three-year project, titled “Collaborative Research: SHF: Hierarchical Optimization of Ragged Tensor Operators for Deep Learning Workloads,” focuses on a type of data structure that does not fit neatly into the fixed-size arrays commonly used by modern machine-learning systems. Unlike regular tensors, whose dimensions are uniform, ragged tensors contain sequences or groups with different lengths. This irregularity is common in real-world AI applications, including natural-language processing, recommendation systems, graph learning, speech recognition and scientific simulations.

Deep-learning frameworks are designed to exploit the massive parallelism of graphics processing units, or GPUs. These processors can execute thousands of operations simultaneously, making them central to the training and deployment of contemporary AI models. However, GPUs achieve their greatest efficiency when data is organized predictably. Ragged tensors disrupt that regularity: one input may contain a few elements, while another may contain thousands. As a result, GPU threads can be forced to wait, memory accesses can become inefficient, and computational resources may sit idle.

Zhou, an assistant professor of computer science in George Mason’s College of Engineering and Computing, and Yin, a professor in the same department, are seeking to address these limitations through a combination of algorithms, data structures, compiler technology and automated performance tuning. Their work will examine the fundamental computational behavior of ragged tensor operators—the routines that manipulate irregular data during AI workloads—and develop methods for making those routines faster, more scalable and easier to integrate into existing software systems.

A central innovation will be a multi-level, multi-phase auto-tuning module designed to adapt computing strategies to dynamic workloads. Auto-tuning allows software to evaluate multiple implementation choices and select the one most likely to deliver high performance for a particular input, GPU architecture or execution environment. For ragged tensors, that decision is especially complicated because the shape and distribution of the data can change from one operation to the next. The researchers’ hierarchical approach is intended to make these decisions at several levels, balancing global workload behavior with the detailed characteristics of individual operations.

The project will also investigate ways to reduce the overhead associated with irregular computation. In a conventional deep-learning pipeline, the time spent preparing data, coordinating GPU threads and moving information between processors can undermine the benefits of hardware acceleration. Zhou and Yin plan to develop mechanisms that overlap computation with communication, allowing useful processing to continue while data is transferred or other system tasks are completed. If successful, the approach could reduce idle time and improve throughput in applications that rely on large, constantly changing datasets.

Another major component is a compiler module that can incorporate ragged tensor operators into established deep-learning workflows. Compilers translate high-level programs into machine instructions, and a specialized compiler layer could automate many of the optimizations currently handled manually by programmers. The proposed module is intended to connect irregular operators with existing frameworks and hardware without requiring researchers or developers to redesign their entire software stack. That compatibility could be critical for bringing efficient ragged-tensor processing from experimental systems into practical AI applications.

The research arrives as the computational demands of artificial intelligence continue to expand beyond conventional image and tabular data. Large language models process sequences that vary in length, recommendation systems analyze users with different histories, and graph-based models work with networks whose nodes may have widely different numbers of connections. In each of these settings, forcing irregular information into uniform structures can require padding, which adds unnecessary memory use and computation. More direct support for ragged data could make AI systems more efficient, particularly when operating under constraints such as limited memory, high energy costs or the need for rapid responses.

The project is scheduled to begin in July 2026 and continue through late June 2029. Its results could influence the design of future AI libraries, GPU kernels and compiler systems by treating irregularity not as an inconvenient exception but as a fundamental feature of modern workloads. By combining adaptive optimization with improved communication and software integration, Zhou and Yin aim to help GPUs handle the messy, uneven data that increasingly defines real-world artificial intelligence. The research is supported by the National Science Foundation and represents George Mason’s broader contribution to high-performance computing and the foundations of next-generation AI.

Subject of Research: Efficient processing and optimization of ragged tensor operators for deep-learning workloads on graphics processing units.

Article Title: Collaborative Research: SHF: Hierarchical Optimization of Ragged Tensor Operators for Deep Learning Workloads

Web References: Mason Now: Power the Possible; George Mason University

Keywords

Ragged tensors, deep learning, artificial intelligence, graphics processing units, GPU optimization, auto-tuning, compiler technology, high-performance computing, irregular data, National Science Foundation, George Mason University

At Rice University, Dr. Tianjun Sun, an assistant professor in psychological sciences, is spearheading a pioneering NSF-funded research project aimed at deconstructing the mechanisms behind AI-driven interviews. This collaborative endeavor with the University of Florida spans two years, targeting the core of AI interview methodologies to uncover potential biases inherent in these complex systems. Sun’s motivation stems from growing evidence indicating that while AI-based tools offer consistency in candidate assessment, they may also inadvertently perpetuate or even amplify discriminatory biases, often tied to gender, ethnicity, and cultural backgrounds.

The research hinges on the fundamental concern that AI algorithms do not interpret human language or behavioral cues neutrally. Two individuals can provide equivalent answers to the same question, but the underlying natural language processing (NLP) models and scoring algorithms may evaluate these responses disparately. This discrepancy emerges due to subtle linguistic markers and speech patterns, which algorithms trained on historical data might misclassify. This phenomenon potentially leads to unfair hiring outcomes, whereby an ostensibly neutral chatbot might unwittingly disadvantage certain demographic groups, violating principles of equity in recruitment.

Understanding the intricacies of these AI systems requires considering the layered architecture of chatbot interviews. Dr. Sun’s project meticulously examines three critical levels: first, the predictors, which involve the linguistic and paralinguistic features the AI extracts from candidates’ responses; second, the outcomes, encompassing the computed scores and recommendation metrics the system generates; and third, the candidates’ perceptions, including how applicants judge the fairness, transparency, and legitimacy of the interview process. This tripartite framework sheds light not only on algorithmic biases but also on human experiences of AI-mediated hiring, thereby broadening the analysis to include psychological impacts.

One innovative aspect of Sun’s research is the development of a prototype AI chatbot capable of conducting concise interviews and producing personality assessments based on the widely recognized Big Five personality traits. This psychometric framework evaluates openness, conscientiousness, extraversion, agreeableness, and neuroticism, providing a multidimensional profile of candidates beyond conventional skill metrics. Such a system claims to blend psychological rigor with AI’s computational power, illustrating a new frontier where data-driven personality analysis might complement traditional evaluation methods to enhance predictive validity and fairness.

Dr. Sun emphasizes a groundbreaking conceptual approach she terms “psychometric AI,” which integrates century-old psychological measurement principles with contemporary algorithmic design. Unlike conventional computer science models that prioritize predictive accuracy or optimization metrics, psychometric AI interrogates whether the system genuinely measures its intended constructs and whether its decision-making process adheres to ethical and equitable standards. This approach challenges the field to move beyond “black box” predictive performance towards transparent, explainable, and socially responsible AI deployment in hiring.

The broader context of this research is underscored by industry trends. A growing number of companies—from startups to multinational corporations—have adopted AI tools to streamline recruitment, leveraging chatbots for initial candidate screening. However, as adoption accelerates, academic and governmental watchdogs have documented instances where algorithmic biases have produced discriminatory hiring outcomes, often reflecting pre-existing societal inequalities embedded in training datasets. These systemic concerns highlight an urgent need for establishing scientific benchmarks that can guide AI development towards fairness and accountability.

Reflecting on the societal implications, Patricia DeLucia, associate dean for research at Rice’s School of Social Sciences, underscores the transformative potential of Sun’s work. As AI permeates more facets of daily life and critical decision-making, having rigorous, interdisciplinary research that anticipates ethical challenges is crucial. DeLucia views this project as emblematic of research that not only advances scientific knowledge but also generates actionable insights with tangible societal benefits, particularly amid growing calls for AI governance frameworks.

From a technical perspective, the AI interview systems under scrutiny employ advanced natural language processing models trained on diverse text corpora. These models extract syntactic, semantic, and stylistic features to analyze candidate responses. However, subtle variations in dialect, phrasing, or cultural expression can confound algorithms, causing inconsistent scoring. Addressing this requires sophisticated bias mitigation techniques, including adversarial training, fairness-aware machine learning, and embedding psychological constructs that are robust across demographic groups.

Furthermore, the transparency of AI decision-making plays a pivotal role in user acceptance. Candidates’ trust in AI-mediated interviews hinges on understanding how their responses influence outcomes. Psychological research shows that perceived fairness and process legitimacy directly influence candidate satisfaction and subsequent engagement, affecting employer brand and recruiting success. Sun’s project, therefore, incorporates perceptual measures designed to capture these applicant attitudes, complementing technical bias assessments with human-centered evaluations.

If successful, the outcomes of this project will set critical precedents and create practical guidelines for constructing AI hiring tools that are not only effective but also equitable, transparent, and psychologically sound. Employers may gain frameworks to audit, diagnose, and recalibrate AI systems, transforming chatbot interviews from opaque automated gatekeepers into tools that enhance diversity and inclusivity in hiring practices.

With AI’s rise as the vanguard of recruitment innovation, this research signals a pivotal step toward reconciling technological efficiency with social justice. The stakes transcend hiring alone, touching on broader societal values of fairness, trust, and opportunity in an increasingly automated world. Through the integration of psychology and computer science, Dr. Sun’s pioneering study illuminates an urgent path forward—designing AI that understands human complexity without sacrificing ethical responsibility.

Subject of Research: Artificial intelligence in automated job interviews and the assessment of fairness and bias in AI hiring tools.

Article Title: Toward Fairer AI Job Interviews: Bridging Psychometrics and Machine Learning in Automated Hiring Systems

News Publication Date: Not specified

Web References: https://www.nsf.gov/

Keywords: Artificial intelligence, AI hiring systems, chatbot interviews, psychometric AI, fairness in AI, personality psychology, Big Five personality traits, natural language processing, bias mitigation, ethical AI, social sciences, psychological measurement