Rimon Melamed

PhD Student

About Me

I’m a co-founder and CTO at Prompt Inversion, an applied AI lab. I’m also a computer science PhD student at The George Washington University advised by H. Howie Huang. My research is broadly in deep learning and natural language processing.

Prior to starting my PhD I was a Software Engineer II at Intelligent Automation where I worked on CryptoniteNXT, a hardware-based on-prem solution for cyber network defense.

Interests

Deep Learning
Natural Language Processing
Robustness
Trustworthiness
Interpretability

Education

PhD in Computer Science, January 2022 - Present
The George Washington University
B.S. in Computer Science, August 2016 - May 2020
University of Maryland, College Park

Selected Publications & Preprints

Rimon Melamed, Lucas H. McCabe, H. Howie Huang (2025). Demystifying optimized prompts in language models. In EMNLP 2025.

PDF Cite Code

Lucas H. McCabe, Rimon Melamed, Thomas Hartivgsen, H. Howie Huang (2025). Estimating Semantic Alphabet Size for LLM Uncertainty Quantification. Preprint, under review.

PDF Cite

Rimon Melamed, Lucas H. McCabe, Tanay Wakhare, Yejin Kim et al. (2024). Prompts have evil twins. In EMNLP 2024 (Oral - top 7.3% of submissions).

PDF Cite Code