Liang Receives Dual Honors for Software Research
By Josh Quicksall
Jenny T. Liang, a PhD student in Software Engineering at Carnegie Mellon University's Software and Societal Systems Department, recently led research teams that earned two prestigious awards for their innovative papers. "NLPositionality: Measuring Positionality and Design Biases in Datasets and Models in NLP" and "A Qualitative Study on the Implementation Design Decisions of Developers" showcase Liang's vital contributions to both technical and non-technical fields.
The first paper, "A Qualitative Study on the Implementation Design Decisions of Developers," received the coveted SIGSOFT Distinguished Paper Award at the International Conference for Software Engineering (ICSE 2023). This study delved into the software design choices made during the implementation process and their implications for education and practice. By conducting surveys and interviews with industry professionals, Liang and her team exemplified the need to refine and teach effective software design through open-ended projects.
The second award, the Outstanding Paper Award at the Association for Computational Linguistics (ACL 2023), was granted to Liang's paper, "NLPositionality: Characterizing Design Biases of Datasets and Models." This study aimed to tackle biases in natural language processing (NLP) systems by examining creators' positionality. Through diverse participant annotations, Liang's team managed to quantify biases in NLP datasets and models. The findings encourage the development of inclusive NLP systems by carefully considering researchers' positionality as well as model and dataset biases, particularly regarding non-binary individuals and non-native English speakers.
Both papers exemplify the significance of Liang's work, which spans from illuminating the software design decision-making process to raising awareness of prevailing biases in NLP models and datasets. In broader terms, her research promotes equitable AI development and accessible education to ensure fair access and performance for a wide range of user populations.
Liang graduated from the University of Washington with a dual degree in Computer Science and Informatics. Currently, she is a rising second-year Ph.D. student at Carnegie Mellon University, advised by Dr. Brad A. Myers. Her broad interests lie at the intersections of software engineering, human-computer interaction, and natural language processing. Presently, she is engaged in studying methods to enhance developers' interactions with code generation tools.
Sources for the papers, available at: