Amirabbas Afzali

Hi, I’m Amirabbas! I am an incoming DPhil student in Engineering Science at the University of Oxford, supervised by Prof. Tim G. J. Rudner (University of Toronto & Vector Institute) and Prof. Philip Torr (University of Oxford), where I am supported by a Technical AI Governance DPhil Studentship from the Oxford Martin AI Governance Initiative. I am also a Research Affiliate at the Vector Institute, working with Tim on AI alignment and preference optimization.

I recently completed my B.Sc. in Electrical Engineering at Sharif University of Technology, specializing in Communication Systems. Previously, I was a research intern at the MLBio Lab at EPFL, working with Prof. Maria Brbić on Weak-to-Strong generalization for preference alignment in large language models.

I’m broadly interested in reliable decision-making in machine learning systems, including trustworthy ML and reinforcement learning, especially where these topics intersect with human–AI alignment. My work spans several research areas, including:

Preference learning and alignment of LLMs
Trustworthy ML and AI safety
Off-policy/asynchronous learning
Generalization and provable guarantees in LLMs

selected publications

NeurIPS

LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders

Borna Khodabandeh^*, Amirabbas Afzali^*, Amirhossein Afsharrad, and 4 more authors

Advances in Neural Information Processing Systems, 2025

arXiv Bib

@article{khodabandeh2025lorelagrangianoptimizedrobustembeddings,
  title = {LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders},
  author = {Khodabandeh, Borna and Afzali, Amirabbas and Afsharrad, Amirhossein and Mousavi, Seyed Shahabeddin and Lall, Sanjay and Amini, Sajjad and Moosavi-Dezfooli, Seyed-Mohsen},
  year = {2025},
  journal = {Advances in Neural Information Processing Systems},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2505.18884},
  equal_contribution = {Borna Khodabandeh and Amirabbas Afzali},
}

ICLR

Aligning Visual Contrastive learning models via Preference Optimization

Amirabbas Afzali^*, Borna Khodabandeh^*, Ali Rasekh, and 3 more authors

International Conference on Learning Representations, 2025

arXiv Bib

@article{afzali2025aligningvisualcontrastivelearning,
  title = {Aligning Visual Contrastive learning models via Preference Optimization},
  author = {Afzali, Amirabbas and Khodabandeh, Borna and Rasekh, Ali and JafariNodeh, Mahyar and kazemi, Sepehr and Gottschalk, Simon},
  year = {2025},
  journal = {International Conference on Learning Representations},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
  url = {https://arxiv.org/abs/2411.08923},
  equal_contribution = {Amirabbas Afzali and Borna Khodabandeh},
}

RLC

One Goal, Many Challenges: Robust Preference Optimization Amid Content-Aware and Multi-Source Noise

Amirabbas Afzali, Amirhossein Afsharrad, Seyed Shahabeddin Mousavi, and 1 more author

Reinforcement Learning Conference, 2025

arXiv Bib

@article{afzali2025goalchallengesrobustpreference,
  title = {One Goal, Many Challenges: Robust Preference Optimization Amid Content-Aware and Multi-Source Noise},
  author = {Afzali, Amirabbas and Afsharrad, Amirhossein and Mousavi, Seyed Shahabeddin and Lall, Sanjay},
  year = {2025},
  journal = {Reinforcement Learning Conference},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2503.12301},
}