MohammadHossein Rezaei

Email: mhrezaei@arizona.edu

I am a Machine Learning Research Engineer at Scale AI where I work on post-training and evaluation of LLMs. I worked on OnlineRubrics, an approach for post-training LLMs with evolving rubrics to improve alignment in tasks without verifiable ground-truth.

I earned a B.S. in Computer Science from the University of Arizona UArizona . I was a member of the Computational Language Understanding (CLU) Lab, advised by Eduardo Blanco, where I worked on making SLMs more robust against negation by further pre-training and paraphrasing in affirmative terms.

Previously, I was a research intern at Stanford University in the SALT Lab advised by Diyi Yang. There, I co-created EgoNormia, a benchmark for evaluating physical-social norm understanding in vision-language models.

news

Jan 05, 2026	I moved to New York City to join Scale AI as a Machine Learning Research Engineer, Post-training.
Dec 19, 2025	I graduated Summa Cum Laude with a B.S. in Computer Science and a Minor in Mathematics. I delivered the keynote address at the College of Science Convocation Ceremony.
Dec 17, 2025	I was selected as the Overall Outstanding Senior for both the Computer Science Department and the College of Science at the University of Arizona.
Oct 09, 2025	Check out my internship project at Scale AI: Online Rubrics Elicitation from Pairwise Comparisons.
May 27, 2025	I joined Scale AI as a Research Intern, Post-training.
May 15, 2025	EgoNormia: Benchmarking Physical Social Norm Understanding has been accepted to ACL 2025 Findings.
Jan 22, 2025	My paper, Making Language Models Robust Against Negation, has been accepted to NAACL 2025. See you in Albuquerque!

selected publications

Scale AI

Online Rubrics Elicitation from Pairwise Comparisons

MohammadHossein Rezaei, Robert Vacareanu, Zihao Wang , Clinton Wang, Bing Liu, Yunzhong He, and Afra Feyza Akyürek

2025

arXiv Bib PDF Website

@misc{rezaei2025onlinerubricselicitationpairwise,
  title = {Online Rubrics Elicitation from Pairwise Comparisons},
  author = {Rezaei, MohammadHossein and Vacareanu, Robert and Wang, Zihao and Wang, Clinton and Liu, Bing and He, Yunzhong and Akyürek, Afra Feyza},
  year = {2025},
  eprint = {2510.07284},
  archiveprefix = {arXiv},
  primaryclass = {cs.CL},
  url = {https://arxiv.org/abs/2510.07284},
}

ACL (Findings)

EgoNormia: Benchmarking Physical-Social Norm Understanding

MohammadHossein Rezaei^*, Yicheng Fu^*, Phil Cuvin^*, Caleb Ziems, Yanzhe Zhang, Hao Zhu, and Diyi Yang

In Findings of the Association for Computational Linguistics: ACL 2025, Jul 2025

DOI arXiv Bib PDF Code Poster Slides Website

@inproceedings{rezaei-etal-2025-egonormia,
  title = {EgoNormia: Benchmarking Physical-Social Norm Understanding},
  author = {Rezaei, MohammadHossein and Fu, Yicheng and Cuvin, Phil and Ziems, Caleb and Zhang, Yanzhe and Zhu, Hao and Yang, Diyi},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2025},
  month = jul,
  year = {2025},
  address = {Vienna, Austria},
  url = {https://aclanthology.org/2025.findings-acl.985/},
  publisher = {Association for Computational Linguistics},
  doi = {10.18653/v1/2025.findings-acl.985},
  pages = {19256--19283},
  isbn = {979-8-89176-256-5},
}

NAACL
Making Language Models Robust Against Negation

MohammadHossein Rezaei, and Eduardo Blanco

In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Apr 2025

Abs arXiv Bib PDF Poster Slides

Negation has been a long-standing challenge for language models.Previous studies have shown that they struggle with negation in many natural language understanding tasks.In this work, we propose a self-supervised method to make language models more robust against negation.We introduce a novel task, Next Sentence Polarity Prediction (NSPP), and a variation of the Next Sentence Prediction (NSP) task.We show that BERT and RoBERTa further pre-trained on our tasks outperform the off-the-shelf versions on nine negation-related benchmarks.Most notably, our pre-training tasks yield between 1.8% and 9.1% improvement on CondaQA, a large question-answering corpus requiring reasoning over negation.
@inproceedings{rezaei-blanco-2025-making, title = {Making Language Models Robust Against Negation}, author = {Rezaei, MohammadHossein and Blanco, Eduardo}, editor = {Chiruzzo, Luis and Ritter, Alan and Wang, Lu}, booktitle = {Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)}, month = apr, year = {2025}, address = {Albuquerque, New Mexico}, publisher = {Association for Computational Linguistics}, url = {https://aclanthology.org/2025.naacl-long.413/}, pages = {8123--8142}, isbn = {979-8-89176-189-6}, }
ACL
Paraphrasing in Affirmative Terms Improves Negation Understanding

MohammadHossein Rezaei, and Eduardo Blanco

In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Aug 2024

Abs DOI arXiv Bib PDF Poster Slides

Negation is a common linguistic phenomenon. Yet language models face challenges with negation in many natural language understanding tasks such as question answering and natural language inference. In this paper, we experiment with seamless strategies that incorporate affirmative interpretations (i.e., paraphrases without negation) to make models more robust against negation. Crucially, our affirmative interpretations are obtained automatically. We show improvements with CondaQA, a large corpus requiring reasoning with negation, and five natural language understanding tasks.
@inproceedings{rezaei-blanco-2024-paraphrasing, title = {Paraphrasing in Affirmative Terms Improves Negation Understanding}, author = {Rezaei, MohammadHossein and Blanco, Eduardo}, editor = {Ku, Lun-Wei and Martins, Andre and Srikumar, Vivek}, booktitle = {Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)}, month = aug, year = {2024}, address = {Bangkok, Thailand}, publisher = {Association for Computational Linguistics}, url = {https://aclanthology.org/2024.acl-short.55/}, doi = {10.18653/v1/2024.acl-short.55}, pages = {602--615}, }