Scholar
Hamish Ivison
Google Scholar ID: JxCXMlkAAAAJ
University of Washington
Citations & Impact
Citations
2,212
H-index
12
i10-index
13
Publications
19
Co-authors
4
list available
Publications
Cited
0
Resume (English only)
Academic Achievements
- Contributed to papers such as 'Generalizing Verifiable Instruction Following' and 'The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains'; also helps maintain the Open-Instruct codebase for general LM post-training.
Research Experience
- Worked on multi-hop question answering with the UsydNLP group; spent time at the Commonwealth Bank of Australia, startups, and Optiver; was a predoctoral researcher at AI2 on the AllenNLP team before his PhD.
Education
- Undergraduate from the University of Sydney, triple majoring in Linguistics, Classical Greek, and Computer Science; currently a PhD student at the University of Washington's H2Lab, advised by Hannaneh Hajishirzi.
Background
- Research interests include NLP, with a focus on post-training for language models, making them more usable and exploring ways to improve them beyond next-token training. Also interested in improving and exploring language model data mixtures and alternative approaches to language modeling.
Miscellany
- From Sydney, happy to answer questions about work, academia, software, or research-related stuff. Can be reached at hamishiv [at] cs [dot] washington [dot] edu.
