Taha Binhuraib

About

I'm a Ph.D. student in Cognition and Brain Sciences at Georgia Tech, where I'm fortunate to work with Anna Ivanova in the LIT Lab. I study the intersection of language processing in the brain and neural networks, focusing on building interpretable models that map language to brain activity and uncovering the algorithmic foundations of Topographic Language Models.

My interest in neuroscience grew out of a deep fascination with philosophy. Early on, I was drawn to questions about the mind, consciousness, and logic, which eventually led me from studying classical thinkers to exploring the neural basis of thought and language (Philosophical Perspectives).

Publications

LITcoder: A General-Purpose Library for Building and Comparing Encoding Models

T Binhuraib, R Gao, AA Ivanova

NeurIPS DBM 2025 (Accepted)

[Paper] [Project Page]

From Language to Cognition: How LLMs Outgrow the Human Language Network

B AlKhamissi, G Tuckute, Y Tang, T Binhuraib, A Bosselut, M Schrimpf

EMNLP 2025 (Oral)

[Paper]

TopoLM: brain-like spatio-functional organization in a topographic language model

N Rathi, J Mehrer, B AlKhamissi, T Binhuraib, NM Blauch, M Schrimpf

ICLR 2025 (Oral)

[Paper] | [Code]

Topoformer: brain-like topographic organization in Transformer language models through spatial querying and reweighting

T Binhuraib, G Tuckute, NM Blauch

ICLR Re-Align 2024

[Poster] | [Project Page]

Protodiffusion: classifier-free diffusion guidance with prototype learning

G Baykal, H F Karagoz, T Binhuraib, and G Unal

ACML 2023

[Paper]

Teaching

• PSYC 2760 - Human Language Processing
  Teaching Assistant, Georgia Tech (Fall 2025)
• PSYC 2760 - Human Language Processing
  Teaching Assistant, Georgia Tech (Fall 2024)

Research

At the Language, Intelligence and Thought (LIT) Lab, I'm working on developing interpretable models for encoding language in the brain. My research interests span across:

• Neural encoding of language in the brain
• Topographic organization in language models
• Interpretable machine learning for neuroscience
• Brain-like architectures in artificial neural networks

Background

I hold a B.Sc. in Industrial Engineering from Bahcesehir University, where I graduated with high honors as the second-ranking student in my department. Following my undergraduate studies, I worked in industry as a Data Scientist and later as a Machine Learning Engineer. I then became the Head of the Machine Learning Engineering team at Novus ASI, where I led the end-to-end development of production-scale Turkish language models and agentic AI applications. In parallel, I also worked as a Researcher at Istanbul Technical University with Gozde Unal. This combination of industry leadership and academic research experience, together with my current focus on neuroscience, gives me a unique perspective on the intersection of artificial and biological intelligence.