Scholar
Shuyue Stella Li
Google Scholar ID: CRfOlOEAAAAJ
University of Washington
Citations & Impact
Citations
368
H-index
8
i10-index
8
Publications
20
Co-authors
15
list available
Publications
Resume (English only)
Academic Achievements
- 2025: Published 'PrefPalette: Personalized Preference Modeling with Latent Attributes' for cognitively grounded preference learning.
- 2025: Published 'Spurious Rewards: Rethinking Training Signals in RLVR', showing performance gains with incorrect rewards.
- 2025: Published 'A False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakage'.
- 2025: Published 'ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning'.
- 2024: Published 'MEDIQ: Question-Asking LLMs for Adaptive and Reliable Clinical Reasoning'.
- 2024: Published 'ValueScope: Unveiling Implicit Norms and Values via Return Potential Model of Social Interactions'.
- Presented MediQ at NeurIPS 2024 (Vancouver).
- Presented ValueScope and Multilingual Abstention at EMNLP 2024 (Miami).
- Gave talks on MediQ at JHU RAIHS Symposium and MSR Real-world Evidence Lab in 2024.
Background
- Second-year Ph.D. student at the Allen School of Computer Science and Engineering, University of Washington, advised by Yulia Tsvetkov.
- Research focuses on Natural Language Processing and Cognitive Modeling, with an emphasis on applying AI to improve healthcare and wellness outcomes.
- Interested in computational modeling of human cognitive processes, including reasoning, communicating uncertainty, and decision-making in complex domains like healthcare.
- Long-term goal: building socially and cognitively aligned NLP systems for safer, more personalized, and equitable care.
- Research interests: AI for Health, Safety & Reliability, Proactive Learning, Social Reasoning, etc.
Miscellany
- Maintains a 'Photography' section on her personal website, indicating interest in photography.
- Email: stelli@cs.washington.edu
- Google Scholar ID: CRfOlOEAAAAJ
- GitHub: stellalisy
- Location: Seattle, Washington, USA
