Scholar
Yi-Hao Peng
Google Scholar ID: yAsgeGIAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
913
H-index
16
i10-index
21
Publications
20
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- Morae: Proactively Pausing UI Agents for User Choices (ACM Symposium on User Interface Software and Technology (UIST 2025))
- StepWrite: Adaptive Planning for Speech-Driven Text Generation (ACM Symposium on User Interface Software and Technology (UIST 2025))
- Towards Bidirectional Human-AI Alignment: A Systematic Review for Clarifications, Framework, and Future Directions (Advances in Neural Information Processing Systems (NeurIPS 2025), Position Paper Track)
- Tempura: Temporal Event Masked Prediction and Understanding for Reasoning in Action (arXiv.2025.05 (under submission))
- CodeA11y: Making AI Coding Assistants Useful for Accessible Web Development (ACM Conference on Human Factors in Computing Systems (CHI 2025))
- AutoPresent: Designing Structured Visuals From Scratch (IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2025))
- DreamStruct: Understanding Slides and User Interfaces via Synthetic Data Generation (European Conference on Computer Vision (ECCV 2024))
- UIClip: A Data-driven Model for Assessing User Interface Design (ACM Symposium on User Interface Software and Technology (UIST 2024))
- Long-form Answers to Visual Questions Asked by Blind and Low Vision People (Conference on Language Modeling (COLM 2024))
- AdvCAPTCHA: Creating Usable and Secure Audio CAPTCHA with Adversarial Machine Learning (NDSS Symposium on Usable Security and Privacy (USEC 2024))
Research Experience
- Reviewed Ph.D. applications for more than thirty students worldwide; organizing the CMU Accessibility Seminar series, inviting speakers to give talks on campus or online.
Education
- Currently a Ph.D. student at CMU HCI Institute, advised by professors Jeff Bigham and Amy Pavel.
Background
- Research interests: Human–Computer Interaction, focusing on interactive and accessible AI agents. Explores how such agents can robustly perceive multimodal context, collaborate with users proactively to preserve user agency, and adapt continually to support creative work in design, education, and beyond.
Miscellany
- Personal interests and other information not provided.
