Scholar
Hadas Orgad
Google Scholar ID: xWntyLkAAAAJ
PhD student, Technion
Citations & Impact
Citations
855
H-index
9
i10-index
9
Publications
14
Co-authors
0
Contact
Publications
Cited
0
Resume (English only)
Academic Achievements
- Published several papers, including 'Inside-out: Hidden Factual Knowledge in LLMs' (COLM 2025), 'LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations' (ICLR 2025), 'MIB: A Mechanistic Interpretability Benchmark' (ICML 2025), and more.
Research Experience
- Currently a Research Fellow at the Kempner Institute for the Study of Natural and Artificial Intelligence at Harvard University. Previously worked at Microsoft for 3.5 years on AI solutions for cloud security and NLP applications for security-related problems.
Education
- Completed B.Sc. and M.Sc. degrees at the Technion – Israel Institute of Technology and Ph.D. under the supervision of Yonatan Belinkov. Selected as a 2023 Apple Scholar in AI/ML.
Background
- Research interests include the internals of AI, particularly how interpretability can be used to improve robustness, safety, and trustworthiness. Research problems involve hallucinations, bias, and unsafe outputs.
Miscellany
- Passionate about AI interpretability and open to connecting with others for brainstorming or collaboration.
Co-authors