About

Hello! My name is Deniz, and I am a final-year PhD candidate in Computer Science advised by Prof. Antoine Bosselut at the NLP lab in EPFL, IC.

Broadly, I am interested in understanding the processes underlying model behavior, how these mechanisms develop during training, and how they ultimately influence model predictions, especially in the context of different languages.

More about my research

More specifically, I study how LLMs acquire, organize, and use internal representations, and how we can reliably understand these processes. My work on model interpretability has explored how knowledge is localized within model parameters [1], how linguistic representations emerge and evolve throughout pretraining [2], and how the methods we use to probe models shape the conclusions we draw about their internal computation [3]. I also studied observable model behavior in real-world settings through a large-scale collaboration that compiled assessments from 50 EPFL courses [4].

Previously, I worked on grounded language and language-guided robot instruction-following with Prof. Stefanie Tellex (H2R) and Prof. Ellie Pavlick (LUNAR) at Brown CS.

Publications

Conference & Journal Papers

Lingua Franca or Probing Artifact? Rethinking Latent Language in Multilingual LLMs
D Bayazit, B AlKhamissi, A Bosselut
Findings of the Association for Computational Linguistics
(EMNLP 2026)
[PDF] [code] [poster]

Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretraining
D Bayazit, A Mueller, A Bosselut
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics
(ACL 2026)
[PDF] [code] [video] [poster]

Discovering Knowledge-Critical Subnetworks in Pretrained Language Models
D Bayazit, N Foroutan, Z Chen, G Weiss, A Bosselut
Proceedings of the Conference on Empirical Methods in Natural Language Processing
(EMNLP 2024)
[PDF] [code] [video] [poster]

Could ChatGPT get an Engineering Degree? Evaluating Higher Education Vulnerability to AI Assistants
B Borges*, N Foroutan*, D Bayazit*, A Sotnikova* et al.
Proceedings of the National Academy of Sciences
(PNAS 2024)
[PDF] [code]

PeaCoK: Persona Commonsense Knowledge for Consistent and Engaging Narratives
S Gao, B Borges, S Oh, D Bayazit, S Kanno, H Wakaki, Y Mitsufuji, A Bosselut
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics
(ACL 2023)
Outstanding Paper Award
[PDF] [code] [video]

Spatial Language Understanding for Object Search in Partially Observed Cityscale Environments
K Zheng, D Bayazit, R Mathew, E Pavlick, S Tellex
2021 International Conference on Robot and Human Interactive Communication
(RO-MAN 2021)
[PDF] [website] [code] [video]

Grounding Language to Landmarks in Arbitrary Outdoor Environments.
M Berg*, D Bayazit*, R Mathew, A Rotter-Aboyoun, E Pavlick, S Tellex
2020 International Conference on Robotics and Automation
(ICRA 2020)
[PDF] [data] [video]

Flight, Camera, Action! Using Natural Language and Mixed Reality to Control a Drone
B Huang, D Bayazit, D Ullman, N Gopalan, S Tellex
2019 International Conference on Robotics and Automation
(ICRA 2019)
[PDF] [code] [data] [video]

Preprints

MEDITRON-70B: Scaling Medical Pretraining for Large Language Models
Z Chen, A Hernández-Cano, A Romanou, A Bonnet, K Matoba, F Salvi, M Pagliardini, S Fan, A Köpf, A Mohtashami, A Sallinen, A Sakhaeirad, V Swamy, I Krawczuk, D Bayazit, A Marmet, S Montariol, M Hartley, M Jaggi, A Bosselut
arXiv preprint
(arXiv 2023)
Top ML Papers of the Week (by dair.ai)
[PDF] [code]

Thesis Work

Generalizing Natural Language Instruction Following to Aerial Robots and Arbitrary Environments
D Bayazit
2020 Brown University - Undergraduate Thesis
[PDF]

TA Work

Semester	Course Code	Name	Professor
Spring 2023, 2024	CS 552	Modern Natural Language Processing	Antoine Bosselut
Fall 2022, 2023, 2024	CS 431	Introduction to Natural Language Processing	Jean-Cédric Chappelier, Martin Rajman, Antoine Bosselut
Spring 2021	CSCI 2952I - CLPS 1850	Language Processing in Humans and Machines	Ellie Pavlick, Roman Feiman
Fall 2019	CSCI 1410	Artificial Intelligence	George Konidaris

Deniz Bayazit About Publications TA Work