Scholar
Tianshi Li
Google Scholar ID: FzBd1YYAAAAJ
Assistant Professor, Northeastern University
Citations & Impact
Citations
1,392
H-index
18
i10-index
26
Publications
20
Co-authors
63
list available
Publications
Resume (English only)
Academic Achievements
- Two papers accepted at CHI 2025
- Best Paper Award at HCOMP 2024 for 'Investigating What Factors Influence Users’ Rating of Harmful Algorithmic Bias and Discrimination'
- Paper 'PrivacyLens' accepted to NeurIPS 2024 Datasets and Benchmarks Track, introducing a novel framework to benchmark unintended privacy leakage in LM agents
- Two papers accepted at CSCW 2025 on secret LLM use and ethics of LLM use in HCI research
- Received a $50K gift grant from Google for human-centered privacy protection in text input
- Awarded NSF SaTC grant ($600K total, $200K personal share) on 'Empathy-Based Privacy Education and Design through Synthetic Persona Data Generation'
- Two papers accepted at CHI’24 on LLM privacy invasion and LLM empowerment for multimodal app development
- CHI’24 Special Interest Group proposal on 'Human-Centered Privacy Research in the Age of Large Language Models' accepted
- Co-authored position paper 'Privacy is Not Just Memorization' (Apr 2025), analyzing 1,322 AI/ML privacy papers and revealing that 92% focus on memorization/chat leaks, only 8% on inference-time privacy
- Co-chairing the 1st Workshop on Human-Centered AI Privacy and Security (HAIPS 2025) at CCS 2025 in Taiwan
Background
- Assistant Professor at Northeastern University, Khoury College of Computer Sciences
- Directs the PEACH (Privacy-Enabling AI and Computer-Human interaction) Lab
- Core faculty member at the Cybersecurity and Privacy Institute, Northeastern University
- Believes privacy sustains human agency, safe exploration, and authentic expression in a connected world
- Focuses on studying and addressing emerging LLM privacy issues from a human-centered perspective
- Broad research interests at the intersection of Human-Computer Interaction (HCI), Privacy, and AI
- Conducts mixed-methods research to understand privacy challenges in stakeholders’ lived experiences and builds systems to measure, model, and address these issues
