Anja Surina

I am a PhD student at EPFL advised by Prof. Caglar Gulcehre. My research focuses on using LLMs for mathematical and scientific discovery, and on understanding the nature of their intelligence.

I have previously interned at Google DeepMind on the AlphaEvolve team, at Mila in Prof. Yoshua Bengio’s lab, at Stanford in Prof. Jure Leskovec’s lab, and at IBM Research in Zurich. I received a Master’s degree in Neural Systems and Computation from ETH Zurich and a Bachelor’s degree in Electrical Engineering from the University of Ljubljana.

News

Nov 13, 2025 Selected for Forbes 30u30 in Slovenia in Science&Tech category.
Oct 08, 2025 Presented Algorithm Discovery With LLMs paper at COLM 2025 in Montreal.
Aug 04, 2025 Started as a Student Researcher at Google DeepMind in London.

Publications

  1. The Future of Facts: Tracing the Factual Generation-Verification Gap
    Tim R. Davidson*, Anja Surina*, and Caglar Gulcehre
    arXiv preprint arXiv:2605.27564, 2026
    PDF Blog Code
  2. Advancing Mathematics Research with AI-Driven Formal Proof Search
    George Tsoukalas, Anton Kovsharov, Sergey Shirobokov, Anja Surina, Moritz Firsching, Gergely Bérczi, Francisco J. R. Ruiz, Arun Suggala, Adam Zsolt Wagner, Eric Wieser, Lei Yu, Aja Huang, Miklós Z. Horváth, Andrew Ferrauiolo, Henryk Michalewski, Codrut Grosu, Thomas Hubert, Matej Balog, Pushmeet Kohli, and Swarat Chaudhuri
    arXiv preprint arXiv:2605.22763, 2026
    PDF Code
  3. An Improved Last-Iterate Convergence Rate for Anchored Gradient Descent Ascent
    Anja Surina, Arun Suggala, George Tsoukalas, Anton Kovsharov, Sergey Shirobokov, Francisco JR Ruiz, Pushmeet Kohli, and Swarat Chaudhuri
    arXiv preprint arXiv:2604.03782, 2026
    PDF
  4. Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning
    Anja Surina, Amin Mansouri, Lars C.P.M. Quaedvlieg, Amal Seddas, Maryna Viazovska, Emmanuel Abbe, and Caglar Gulcehre
    COLM, 2025
    PDF Blog Code
  5. The Role of Deep Learning Regularizations on Actors in Offline RL
    Denis Tarasov, Anja Surina, and Caglar Gulcehre
    arXiv preprint arXiv:2409.07606, 2024
    PDF Code
  6. Zero-shot causal learning
    Hamed Nilforoshan, Michael Moor, Yusuf Roohani, Yining Chen, Anja Surina, Michihiro Yasunaga, Sara Oblak, and Jure Leskovec
    NeurIPS, 2023
    Spotlight PDF Code
  7. Biologically-inspired training of spiking recurrent neural networks with neuromorphic hardware
    Thomas Bohnstingl, Anja Surina, Maxime Fabre, Yiğit Demirağ, Charlotte Frenkel, Melika Payvand, Giacomo Indiveri, and Angeliki Pantazi
    2022 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), 2022
    PDF