Kian Ahrabian

PhD Candidate @ University of Southern California (USC)

prof_pic.jpg

Viterbi School of Engineering

University of Southern California

Los Angeles, CA, USA

I am a PhD candidate at the University of Southern California (USC), advised by Jay Pujara. Broadly, my research interests are at the intersection of AI, NLP, and Machine Learning. In particular, I am excited about improving the reasoning capabilities of AI models for textual data and beyond.

Previously, I received a M.Sc. degree in Computer Science from McGill University/MILA and a B.Sc. degree in Computer Science from the University of Tehran. I have also interned at Microsoft Turing (3x), Huawei Noah’s Ark Lab, and Salesforce.

news

Aug 20, 2025 Our paper, “A Systematic Analysis of Base Model Choice for Reward Modeling”, is accepted to EMNLP 2025.
May 19, 2025 Started my internship at Microsoft Turing working on improving the efficiency of reasoning language models.
Mar 31, 2025 I passed my qualifying exam and officially became a PhD Candidate.
Jan 22, 2025 Our paper, “A Practical Analysis of Human Alignment with *PO”, is accepted to NAACL 2025 Findings.
Sep 26, 2024 Our paper, “MARVEL: Multidimensional Abstraction and Reasoning through Visual Evaluation and Learning”, is accepted to NeurIPS 2024 Datasets and Benchmarks Track.

selected publications

  1. EMNLP
    Temporal Knowledge Graph Forecasting Without Knowledge Using In-Context Learning
    Kian Ahrabian*, Dong-Ho Lee*, Woojeong Jin, Fred Morstatter, and Jay Pujara
    In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Dec 2023
  2. COLM
    The Curious Case of Nonverbal Abstract Reasoning with Multi-Modal Large Language Models
    Kian Ahrabian*, Zhivar Sourati*, Kexuan Sun*, Jiarui Zhang, Yifan Jiang, Fred Morstatter, and Jay Pujara
    In First Conference on Language Modeling, Dec 2024
  3. NeurIPS
    MARVEL: Multidimensional Abstraction and Reasoning through Visual Evaluation and Learning
    Yifan Jiang*, Jiarui Zhang*, Kexuan Sun*, Zhivar Sourati, Kian Ahrabian, Kaixin Ma, Filip Ilievski, and Jay Pujara
    In The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track, Dec 2024
  4. LREC-COLING
    On the Adaptation of Unlimiformer for Decoder-Only Transformers
    Kian Ahrabian, Alon Benhaim, Barun Patra, Jay Pujara, Saksham Singhal, and Xia Song
    In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024
  5. NAACL
    A Practical Analysis of Human Alignment with *PO
    Kian Ahrabian, Xihui Lin, Barun Patra, Vishrav Chaudhary, Alon Benhaim, Jay Pujara, and Xia Song
    In Findings of the Association for Computational Linguistics: NAACL 2025, Apr 2025
  6. EMNLP
    A Systematic Analysis of Base Model Choice for Reward Modeling
    Kian Ahrabian, Pegah Jandaghi, Negar Mokhberian, Sai Praneeth Karimireddy, and Jay Pujara
    In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, Nov 2025