Menu

PhD Student

Tom Zehle

Postal address

Institut für Informatik
Albert-Ludwigs-Universität Freiburg
Sekretariat Hutter/Maschinelles Lernen
Georges-Köhler-Allee 074
79110 Freiburg, Germany

Office

Building 074, Room 00-014
EmailLinkedInGoogleScholarORCIDGitHub

Hey, I’m Tom, a PhD Candidate at the ELLIS Institute in Tübingen and a member of the AutoML Lab in Freiburg, supervised by Frank Hutter. My main research interests are Agentic Systems, Prompt Optimization, and Fine Tuning of Foundation Models.

I finished my Masters in Statistics and Data Science at the LMU in 2025, and worked for Airbus from 2020 to 2025 as a Dual Student and Data Scientist.

I coauthored the Prompt Optimization Algorithm CAPO, as well as the Prompt Optimization Library Promptolution.

Publications

2026

Zehle, Tom; Heiß, Timo; Schlager, Moritz; Aßenmacher, Matthias; Feurer, Matthias

promptolution: A Unified, Modular Framework for Prompt Optimization Proceedings Article

In: Croce, Danilo; Leidner, Jochen; Moosavi, Nafise Sadat (Ed.): Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations), pp. 282–296, Association for Computational Linguistics, Rabat, Marocco, 2026, ISBN: 979-8-89176-382-1.

Zehle, Tom; Aßenmacher, Matthias

Can Calibration of Positional Encodings Enhance Long Context Utilization? Proceedings Article

In: Demberg, Vera; Inui, Kentaro; Marquez, Lluís (Ed.): Findings of the Association for Computational Linguistics: EACL 2026, pp. 2268–2280, Association for Computational Linguistics, Rabat, Morocco, 2026, ISBN: 979-8-89176-386-9.

2025

Zehle, Tom; Schlager, Moritz; Heiß, Timo; Feurer, Matthias

CAPO: Cost-Aware Prompt Optimization Proceedings Article

In: AutoML 2025 Methods Track, 2025.