Scholar
John X. Morris
Google Scholar ID: Utsbve4AAAAJ
Cornell University
Citations & Impact
Citations
1,911
H-index
13
i10-index
15
Publications
20
Co-authors
14
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Proposed the first exact measurement of language model capacity and estimated GPT models memorize 3.6 or so bits-per-parameter; built state-of-the-art text embedding model; developed the first method capable of perfectly inverting text embeddings and a follow-up which allowed to invert LLM logits back to prompts; created TextAttack, the first software package for LLM jailbreaking, as an undergraduate in 2019 and 2020.
Research Experience
- A part-time student researcher at Meta (FAIR) for about a third of his PhD; previously a Google Brain Resident.
Education
- PhD from Cornell University (Cornell Tech campus), advisors were Sasha Rush and Vitaly Shmatikov; undergraduate degree from the University of Virginia, specializing in NLP.
Background
- Research interests include Natural Language Processing (NLP) and Machine Learning. During his PhD, he worked closely with Sasha Rush and Vitaly Shmatikov.
Miscellany
- Active on Twitter at jxmnop and GitHub at jxmorris12; appeared on podcasts Latent Space and Odd Lots; started a more formal blog on Substack.
