Hanjie Chen has joined Rice University’s Department of Computer Science as an assistant professor specializing in Natural Language Processing (NLP), Interpretable Machine Learning, and Trustworthy AI. These areas are relatively new within the department, and Chen hopes to bring “fresh perspectives and innovative ideas to the existing curriculum.”
Prior to joining Rice, Chen was a Postdoctoral Fellow in the Center for Language and Speech Processing at Johns Hopkins University, hosted by Mark Drezde. She completed her Ph.D. in Computer Science at the University of Virginia, advised by Yangfeng Ji.
Chen’s current research focuses on “explaining and analyzing the inner workings, reasoning, and prediction processes of neural language models, with a particular emphasis on large language models (LLMs). This includes exploring their emergent capabilities, such as reasoning, text generation, and few-shot learning, as well as identifying their pitfalls and limitations, including biases, adversarial vulnerabilities, and hallucinations.”
Natural Language Processing has had a profound impact on many aspects of modern life, with its applications spanning numerous industries and transforming the way we interact with technology. “For example,” Chen says, “NLP powers virtual assistants like Siri, Alexa, and Google Assistant, which can understand and respond to commands in natural language, enabling people to interact with their devices. These intelligent systems are built on neural language models.”
However, Chen explains, “despite their powerful capabilities in understanding, processing, and generating natural language, they operate in ways that are not fully transparent or explainable to users, developers, or researchers, raising issues of trustworthiness.”
For example, in healthcare, a trustworthy AI system used for diagnostics would need to be “transparent about how it reaches its conclusions, fair in treating patients of different backgrounds, accountable for its decisions, and secure in handling patient data,” Chen points out. “It would also need to be reliable in various medical contexts to ensure it truly benefits patients.”
Chen is working on developing solutions to address these challenges with the aim to build more reliable and trustworthy models for real-world applications. “I’m eager to explore interdisciplinary applications with other researchers and with students,” Chen says. “For example, we could work on developing explainable LLMs for complex medical diagnoses or analyzing model decision-making through a psychological lens.”
She is also affiliated with Rice’s Ken Kennedy Institute, an interdisciplinary group that works collaboratively on groundbreaking research in artificial intelligence, data, and computing. The Ken Kennedy Institute comprises over 250 renowned faculty and senior research members spanning multiple departments and schools. Chen says she is excited to “engage with the Institute's diverse initiatives, such as its flagship conferences and collaborative efforts with external partners like the Texas Medical Center.”
There were many reasons Chen was drawn to Rice: its academic excellence and rigorous programs and its commitment to high quality education and research. A major draw was the excellent computer science faculty and researchers who are doing innovative research across various areas. She looks forward to collaborating with them and contributing to groundbreaking research.
“More importantly,” she says, “the department is welcoming and supportive. Working in a collaborative atmosphere where faculty, staff, and students are approachable and willing to help is something I truly value.”