I’m a first-year PhD candidate in the Schwartslab of Huji-NLP at the Hebrew University. I’m currently on a full-year research visit at DILL Lab with Swabha Swayamdipta.

I'm broadly interested in how language models form and organize meaning internally. My research revolves around the inner lexicon: the emergent word-level representations LLMs build from sub-word fragments, and how understanding this hidden structure can make models more faithful, efficient, and interpretable.

When I’m not delving into cutting-edge NLP challenges, I’m a passionate Tel Aviv fan, a devoted cat parent, a sea lover, and arguably the most enthusiastic reader around.

Tel Aviv Art Decal

News

Publications

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models

ACL 2026

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models

Guy Kaplan*, Michael Toker*, Yuval Reif, Yonatan Belinkov, Roy Schwartz (*equal contribution)

Vocab Diet: Reshaping the Vocabularies of LLMs with Vector Arithmetic

ACL Findings 2026

Vocab Diet: Reshaping the Vocabularies of LLMs with Vector Arithmetic

Yuval Reif, Guy Kaplan, Roy Schwartz

From Tokens to Words: On the Inner Lexicon of LLMs

ICLR 2025

From Tokens to Words: On the Inner Lexicon of LLMs

Guy Kaplan, Matanel Oren, Yuval Reif, Roy Schwartz

Preprints

SFT-Induced Hallucinations as a Continual Learning Problem

SFT-Induced Hallucinations as a Continual Learning Problem

Guy Kaplan, Zorik Gekhman, Zhen Zhu, Yuval Reif, Lotem Rozner, Swabha Swayamdipta, Derek Hoiem, Roy Schwartz

Submitted to CoLM 2026

More Than Words: Compositional Tokenization for Efficient Language Models

More Than Words: Compositional Tokenization for Efficient Language Models

Yuval Reif, Guy Kaplan, Roy Schwartz

Submitted to CoLM 2026

Education

Aug. 2025 – Present

The Hebrew University of Jerusalem

PhD Student, Computer Science

Advisor: Prof. Roy Schwartz

Oct. 2022 – Aug. 2025

The Hebrew University of Jerusalem

M.Sc., Computer Science

Advisor: Prof. Roy Schwartz

Thesis: The Inner Lexicon of Large Language Models

Oct. 2019 – Jun. 2022

The Open University of Israel

B.Sc., Computer Science

Oct. 2019 – Jun. 2022

Tel Aviv University

B.A., Economics

2014

University of Haifa

B.A., General Studies, Magna Cum Laude

Experience

Sep. 2025 – Present
Oct. 2023 – Present

Chief Scientist Officer, Bright Forensic Innovations

Mar. 2024 – Jan. 2025

Data Scientist, Microsoft

Mar. 2022 – Mar. 2024

Software Engineer, Microsoft

Oct. 2011 – Sep. 2019

Naval Officer (Major), Israeli Navy, Air Patrol Unit