Yujia Bao


🧑🏻‍🔬 Advanced AI Research Scientist @Accenture
🐾 A proud owner of samoyeds
🏛 Completed Ph.D. at MIT CSAIL
🏃‍♂️ Staying active in Cambridge, MA

Hello! I’m a machine learning researcher with a background in Mathematics. I’ve been immersed in this field since 2016 and have had the opportunity to work on some fascinating projects.

Developing NLP Systems with Minimal Task-Dependent Annotations

I’ve focused on creating more efficient NLP systems that require fewer task-specific annotations. My key contributions include:

  • Enhancing data efficiency by aligning human rationales with attention models.1
  • Constructing distributional signatures for few-shot text classification.2
Addressing Bias in Machine Learning

I firmly believe that fairness in machine learning is crucial, especially as biased datasets can lead to biased models. My work in this area includes:

  • Mitigating spurious correlations by contrasting different data environments.3
  • Transferring the knowledge of biases across tasks.4
  • Automatic bias discovery by learning challening splits from labeled datasets.5
Advancing Foundational Models for Cell Imaging

In the realm of cell imaging, I’ve been at the forefront of developing foundational models. My work includes:

  • Enhancing vision transformers with context pre-conditioning for improved generalization over distribution shifts.6
  • Enabling cross-channel and cross-position reasoning for vision transformers on multi-channel imaging.7

I’m driven by my love for learning and the excitement that comes from creating something new. Each day offers a new chance to explore and innovate, which I find truly thrilling.


🎓 B.S. in Mathematics & Applied Mathematics, Shanghai Jiao Tong University, 2016.
🎓 M.A. in Mathematics, University of Wisconsin-Madison, 2017.
🎓 S.M. in Computer Science, MIT, 2019.
🎓 Ph.D. in Computer Science, MIT, 2022 (advised by Regina Barzilay).


  1. Yujia Bao, Shiyu Chang, Mo Yu, Regina Barzilay. “Deriving Machine Attention from Human Rationales.” EMNLP 2018.
  2. Yujia bao, Menghua Wu, Shiyu Chang, Regina Barzilay. “Few-shot Text Classification with Distributional Signatures.” ICLR 2020.
  3. Yujia Bao, Shiyu Chang, Regina Barzilay. “Predict then Interpolate: A Simple Algorithm to Learn Stable Classifiers.” ICML 2021.
  4. Yujia Bao, Shiyu Chang, Regina Barzilay. “Learning Stable Classifiers by Transferring Unstable Features.” ICML 2022.
  5. Yujia Bao, Regina Barzilay. “Learning to Split for Automatic Bias Detection.” arXiv 2022.
  6. Yujia Bao, Theofanis Karaletsos. “Contextual Vision Transformers for Robust Representation Learning” SCIS Workshop at ICML 2023.
  7. Yujia Bao, Srinivasan Sivanandan, Theofanis Karaletsos. “Channel Vision Transformers: An Image Is Worth C x 16 x 16 Words” UniReps Workshop at NeurIPS 2023.