PhD Researcher · JKU Linz

Christoph
Bartmann

At the intersection of machine learning and chemistry — building reasoning systems that think about molecules the way scientists do.

scroll
about.sh
$ cat --about christoph.yml
PhD researcher at the Institute for Machine Learning, JKU Linz, supervised by Sepp Hochreiter and Günter Klambauer. $ grep --research interests
My work sits at the crossroads of reinforcement learning and scientific reasoning. I build RL-based pipelines that teach LLMs to reason about molecules, evaluate chemical understanding through symbolic verification, and develop reproducible benchmarks for AI in drug discovery. $ echo $BACKGROUND
Two MSc degrees — Data Science and Chemistry — from TU Vienna, with a research stay at UC Berkeley as a Marshall Plan Fellow. I think the best AI for science comes from people who speak both languages. # currently building: reasoning pipelines with TRL, VERL, NeMo-RL

Papers

03
ICLR 2026
MolecularIQ: Characterizing Chemical Reasoning Capabilities Through Symbolic Verification on Molecular Graphs
C. Bartmann, J. Schimunek, M. Ielanskyi, P. Seidl, G. Klambauer, S. Luukkonen
NeurIPS 2025
Rethinking Losses for Diffusion Bridge Samplers
S. Sanokowski, L. Gruber, C. Bartmann, S. Hochreiter, S. Lehner
arXiv 2025
Measuring AI Progress in Drug Discovery: A Reproducible Leaderboard for the Tox21 Challenge
A. Ebner*, C. Bartmann*, S. Topf, S. Luukkonen, J. Schimunek, G. Klambauer

Lab Notes

00
$ echo "posts incoming..."
// lab notes coming soon