Scholar
Lei Hsiung
Google Scholar ID: CJaZ2NcAAAAJ
Dartmouth College
Citations & Impact
Citations
116
H-index
5
i10-index
3
Publications
12
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'NeuralFuse' accepted to NeurIPS 2024
- Paper 'AutoVP' accepted to ICLR 2024 (acceptance rate: 30.98%)
- Paper 'Data Debiasing via Model-free Data Pruning' accepted to ICLR 2024 Workshop on Navigating and Addressing Data Problems for Foundation Models
- Recipient of ICLR 2024 Scholar Award
- Paper 'Spectral Insights into Data-Oblivious Critical Layers in Large Language Models' accepted to Findings of ACL 2025
- Paper 'Why LLM Safety Guardrails Collapse After Fine-tuning' under submission
- Selected for OpenAI Researcher Access Program (June 2025)
- Served as reviewer for NeurIPS 2024/2025, ICML 2025, ICLR 2024, AAAI 2025
Background
- Second-year Ph.D. student in Computer Science at Dartmouth College
- Research lies at the intersection of trustworthy and efficient machine learning
- Focuses on building safer and more reliable machine learning models
- Research topics include safety alignment, adversarial robustness, model reprogramming, and energy-efficient inference
- Aims to advance the safety and reliability of ML systems as a foundation for robust and secure artificial general intelligence
