Dora Zhao

About Dora Zhao

Dora Zhao was born in Newtown, Pennsylvania to Chinese immigrants. Her parents had moved from China to Japan for graduate school before immigrating to the United States in search of further educational and professional opportunities. From a young age, she was enamored with technology's capability to connect people, from video-calling family members still in China to making friends via online communities. However, she also became aware of the ways in which these technologies could cause harm. Most saliently, she observed how devices, such as Siri or Alexa, routinely failed when her parents used them. These experiences spurred her to study computer science to help create technologies that serve diverse communities and amplify marginalized voices.

Dora graduated from Princeton University summa cum laude with a bachelor’s in computer science and certificates in Asian American studies and statistics & machine learning. She also completed her master’s in computer science at Princeton University. 

During her undergraduate degree, Dora deepened her interest in building more inclusive technical systems through conducting research on fairness and bias in computer vision with Professor Olga Russakovsky. In the lab, she led efforts to audit racial biases in image captioning models and contributed to the creation of an object recognition dataset sourced globally, emphasizing images from the Global South. This experience inspired Dora to examine broader challenges in dataset collection, leading to research on improving dataset reliability and validity through frameworks grounded in measurement theory. This work was recognized with a best paper award at the International Conference on Machine Learning. Dora's research has also led to several other first-author publications at major machine learning conferences, including Advances in Neural Information Processing Systems and the International Conference on Computer Vision.

Currently, Dora is pursuing a PhD in computer science at Stanford University where she is co-advised by Professors Michael Bernstein and Diyi Yang. Her work focuses on building technical systems that give communities control over the design and creation of large language models. Through this research, she hopes to help build a future for artificial intelligence in which all individuals and communities have a say over how these technologies are made and deployed.