About Me
I'm a student at the University of Oxford, working with Haggai Maron, Michael Bronstein, and Yarin Gal. I am broadly interested in language model architectures, long horizon tasks, and the structure of neural network parameter spaces. I am currently an intern at Meta, previously spent time at Sakana AI, and co-organized the Weight Space Learning Workshop at ICLR 2025. Before Oxford, I studied at the Technion as a Rothschild scholar and taught math and programming at the ARDC.
Selected Publications
* denotes equal contribution; full list on Google Scholar
Training Transformers for KV Cache Compressibility
*, Yam Eitan*, Michael Bronstein, Yarin Gal, Haggai Maron
HiLD @ ICML 2026
Extending the Context of Pretrained LLMs by Dropping Their Positional Embeddings
, Koshi Eguchi, Takuya Akiva, Edoardo Cetin
ICLR 2026
GradMetaNet: An Equivariant Architecture for Learning on Gradients
*, Yam Eitan*, Aviv Navon, Aviv Shamsian, Theo (Moe) Putterman, Michael Bronstein, Haggai Maron
NeurIPS 2025
Weight Space Learning Workshop @ ICLR 2025
Leaning on LoRAs: GL-Equivariant Processing of Low-Rank Weight Spaces for Large Finetuned Models
Theo (Moe) Putterman*, Derek Lim*, , Stephanie Jegelka, Haggai Maron
LoG 2025, Oral Presentation 🎤
Weight Space Learning Workshop @ ICLR 2025
Misalignment Between Vision-Language Representations in Vision-Language Models
Yonatan Gideoni, , Tim G. J. Rudner, Yarin Gal
UniReps + CogInterp Workshops @ NeurIPS 2025
Beyond Next Token Probabilities: Learnable, Fast Detection of Hallucinations and Data Contamination on LLM Output Distributions
Fabrizio Frasca*, Guy Bar-Shalom*, Derek Lim, , Yftah Ziser, Ran El-Yaniv, Gal Chechik, Haggai Maron
AAAI 2026
R2-FM Workshop @ ICML 2025
Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity
Yam Eitan*, *, Guy Bar-Shalom, Fabrizio Frasca, Michael Bronstein, Haggai Maron
ICLR 2025, Oral Presentation 🎤
Variational Inference Failures Under Model Symmetries: Permutation Invariant Posteriors for Bayesian Neural Networks
, Tycho van der Ouderaa, Mark van der Wilk, Yarin Gal
PMLR, GRaM Workshop @ ICML 2024, Best Paper Award 🏆