A Memory Efficient Unified Algorithm for Online Learning of Linear Dynamical Systems
* Equal contribution
Preprint (under review)
Hello! I'm currently a second-year PhD student at Tel Aviv University, working under the guidance of Prof. Nadav Cohen. I also had the pleasure of being a visiting student researcher at Princeton University, hosted by Prof. Elad Hazan.
My passion lies in the theoretical exploration of fundamental principles underlying deep learning models.
Beyond academics, I enjoy meditation 🧘♂️, boxing 🥊, and mixology 🍸.
Short Bio
2024 - Present
2022 - 2024
2018 - 2021
B.Sc. in Mathematics
Technion – Israel Institute of Technology
Participated in the Technion's excellence program (Rothschild program)
Publications
Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data
Preprint (under review)
Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks
Yuval Ran-Milo
ACL 2026
A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation
Yuval Ran-Milo,Hila Ofek,Shahar Mendel
ACL 2026
Provable Benefits of Complex Parameterizations for Structured State Space Models
NeurIPS 2024
Mamba Knockout for Unraveling Factual Information Flow
* Equal contribution
ACL 2025 (SAC Highlight - top ~5%)
Do Neural Networks Need Gradient Descent to Generalize? A Theoretical Study
NeurIPS 2025
A Generative AI-Empowered Digital Tutor for Higher Education Courses
Hila Reicher,Yarden Frenkel,Maor Juliet Lavi,Rami Nasser,Yuval Ran-Milo,Ron Sheinin,Mark Shtaif,Tova Milo
Information 2025