Greetings! I am Mohammad Reza Ghasemi Madani (Reza), a first-year Ph.D. student at the University of Melbourne, supervised by Dr Caren Han and Dr Jey Han Lau. I was previously a researcher at the SML Group at the University of Trento, working with Dr. Andrea Passerini and Dr. Pasquale Minervini on trustworthy NLP. My primary research focuses include Explainability, Robustness, and Truthfulness of Multimodal Language Models. I did my master’s under the supervision of Dr. Pasquale Minervini (University of Edinburgh) and Dr. Paolo Torroni from the University of Bologna, Italy.

Latest news

• [01/08/2025] I’ve officially started my PhD at the University of Melbourne. Looking forward to doing cool work!

• [08/07/2025] Our latest paper got accepted into CoLM’25. Check it out: Noiser: Bounded Input Perturbations for Attributing Large Language Models. AND it’s my birthday 🎉.

• [02/04/2025] I got admitted to the University of Melbourne PhD in Artificial Intelligence

• [22/01/2025] Are We Done with MMLU? has been accepted to NAACL’25. Yeay!!!

• [08/07/2024] I Just got my first citations. I feel so proud of myself :)

• [07/06/2024] Check out our paper Are We Done with MMLU?

• [01/04/2024] I joined the SML Group as a researcher working with Prof. Andrea Passerini at the University of Trento.

• [02/02/2024] I defended my master’s thesis, “Developing and Comparing Machine Reasoning Models to Humans in NLP Tasks”, with a final grade of 107/110.

• [02/11/2023] I will be presenting REFER: An End-to-end Rationale Extraction Framework for Explanation Regularization at CoNLL in Singapore on December 6-7.

• [19/10/2023] Our paper REFER: An End-to-end Rationale Extraction Framework for Explanation Regularization has been accepted to CoNLL 2023.

Recent publications

• Noiser: Bounded Input Perturbations for Attributing Large Language Models
  Mohammad Reza Ghasemi Madani, Aryo Pradipta Gema, Gabriele Sarti, Yu Zhao, Pasquale Minervini, Andrea Passerini
  Conference on Language Modeling (CoLM) 2025

• Are We Done with MMLU?
  Aryo Pradipta Gema, Joshua Ong Jun Leang, Giwon Hong, Alessio Devoto, Alberto Carlo Maria Mancino, Rohit Saxena, Xuanli He, Yu Zhao, Xiaotang Du, Mohammad Reza Ghasemi Madani, Claire Barale, Robert McHardy, Joshua Harris, Jean Kaddour, Emile van Krieken, Pasquale Minervini
  The Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL) 2025

• REFER: An End-to-end Rationale Extraction Framework for Explanation Regularization
  Mohammad Reza Ghasemi Madani, Pasquale Minervini
  The SIGNLL Conference on Computational Natural Language Learning (CoNLL) 2023

Reviews

• ICLR 2026
• CoNLL 2025
• ICLR 2025
• CoNLL 2024

Contact

Institutional email: reza.ghasemimadani@student.unimelb.edu.au
Personal email: rezamadani.ai@gmail.com