UNIVERSITY PARK, Pa. — Rui Zhang, assistant professor of computer science in the School of Electrical Engineering and Computer Science in the Penn State College of Engineering, earned a five-year, $546,000 U.S. National Science Foundation (NSF) Early Career Development (CAREER) Award for a project titled “Trustworthy Human-Centered Summarization.”
Q: What do you want to understand or solve through this project?
Zhang: The rapid advancement of generative artificial intelligence (AI) and large language models, such as ChatGPT, are transforming how we process data, capture important information, acquire knowledge and make decisions. The goal of this project is to advance the trustworthy summarization of data by centering the design, development and deployment on humans in light of user preferences for controllability, social perspectives for fairness and human knowledge for factuality. For example, using AI to summarize legal documents improves efficiency for legal practitioners to write legal summaries and the public to understand legal events. This requires a trustworthy summarization system by customizing summaries for either the public or the legal practitioner, ensuring fair and unbiased summaries for diverse opinions from different social groups, and guaranteeing truthful outputs with respect to legal knowledge.
To do so, we will establish a comprehensive set of infrastructure including algorithmic foundations, technical innovations, public benchmarks and integrated platforms. Through these innovations, we will advance usable, trustworthy, responsible and safe AI that operates under user control, reflects social norms and honors human knowledge.
Q: How will advances in this area impact society?
Zhang: Our research adopts a human-centered approach to advance alignments of large language models with humans by incorporating user needs of styles and contents, mitigating model bias by integrating diverse values and perspectives, and ensuring factuality with respect to human knowledge and truth.
Through developing the ability to summarize legal documents, scientific research papers, policy reports and clinical text, our work has the potential to impact society long-term by promoting democracy, legal literacy, public scientific literacy and public health.
Q: Will undergraduate or graduate students contribute to this research? How?
Zhang: Yes, both graduate and undergraduate students will contribute to our research. Graduate students will lead the research effort to create next-generation trustworthy summarization systems by establishing rigorous definitions of new tasks, reliable metrics and innovative models. Undergraduate students will participate in data annotation and model evaluation, as well as be involved in education and outreach activities supported by project research outcomes to involve, mentor and empower underrepresented students.
Q: The NSF CAREER award not only funds a research project, but it also recognizes the potential of the recipient as a researcher, educator and leader in their field. How do you hope to fulfill that potential?
Zhang: Deeply honored by this recognition, I am committed to fulfilling the potential this award represents. As a researcher, I plan to advance the field of human-centered summarization, developing innovative techniques that enhance trust and effectiveness in AI-generated summaries. As an educator, I aim to inspire and mentor the next generation of computer scientists by integrating my research into curricula and developing outreach programs for K-12 students. In terms of leadership, I will focus on democratizing AI by making advanced summarization technologies more accessible and understandable to a broader audience, including non-technical users and underserved communities. Ultimately, I hope to contribute to shaping the future of AI in a way that is ethical, trustworthy and beneficial to society, leveraging this CAREER award as a platform to make a lasting impact in my field and beyond.