Veronika Thost

Researcher, Molecular AI, Novo Nordisk

Joined

July 2019

Names

Veronika Thost

Emails

****@ibm.com (Confirmed)

****@gmail.com (Confirmed)

****@zurich.ibm.com (Confirmed)

****@mit.edu (Confirmed)

****@tu-dresden.de

****@novonordisk.com (Confirmed)

Personal Links

Google Scholar

DBLP

ORCID

Career & Education History

Researcher

Molecular AI, Novo Nordisk (novonordisk.com)

2026 – Present

Researcher

AI, IBM Research (ibm.com)

2018 – 2025

Postdoc

TU Dresden (tu-dresden.de)

2017 – 2018

PhD student

TU Dresden (tu-dresden.de)

2012 – 2017

Advisors, Relations & Conflicts

PhD Advisor

Franz Baader

2012 – 2017

Expertise

No areas of expertise listed

Publications

Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification
Kimia Hamidieh, Veronika Thost, Walter Gerych, Mikhail Yurochkin, Marzyeh Ghassemi
- ICLR 2026 Poster
- Readers: Everyone
Uncertainty-Aware LLM Probing
Andrey Bryutkin, Angela Li, Joshua DeOliveira, Elke Rundensteiner, Walter Gerych, Veronika Thost
- Submitted to ICLR 2026
- Readers: Everyone
GPTOpt: Towards Efficient LLM-based Black-Box Optimization
Jamison Meindl, Yunsheng Tian, Tony Cui, Veronika Thost, Zhang-Wei Hong, Jie Chen, Wojciech Matusik, Mina Konakovic Lukovic
- Submitted to ICLR 2026
- Readers: Everyone
Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification
Kimia Hamidieh, Veronika Thost, Walter Gerych, Mikhail Yurochkin, Marzyeh Ghassemi
- NeurIPS 2025 - Reliable ML Workshop
- Readers: Everyone
ZeroShotOpt: Towards Zero-Shot Pretrained Models for Efficient Black-Box Optimization
Jamison Meindl, Yunsheng Tian, Tony Cui, Veronika Thost, Zhang-Wei Hong, Johannes P. Dürholt, Jie Chen, Wojciech Matusik, Mina Konakovic Lukovic
- ScaleOPT Poster
- Readers: Everyone
Self-Supervised Learning on Molecular Graphs: A Systematic Investigation of Masking Design
Jiannan Yang, Veronika Thost, Tengfei Ma
- Accepted by TMLR
- Readers: Everyone
Do Sparse Autoencoders Generalize? A Case Study of Answerability
Lovis Heindrich, Philip Torr, Fazl Barez, Veronika Thost
- ICML 2025 R2-FM Workshop Poster
- Readers: Everyone
Masking in Molecular Graphs Leveraging Reaction Context
Jiannan Yang, Veronika Thost, Tengfei Ma
- ICML2024-AI4Science Poster
- Readers: Everyone
Knowledge Graph Compression Enhances Diverse Commonsense Generation
EunJeong Hwang, Veronika Thost, Vered Shwartz, Tengfei Ma
- 04 Feb 2024
- OpenReview Archive Direct Upload
- Readers: Everyone
Representing Molecules as Random Walks Over Interpretable Grammars
Michael Sun, Minghao Guo, Weize Yuan, Veronika Thost, Crystal Elaine Owens, Aristotle Franklin Grosz, Sharvaa Selvan, Katelyn Zhou, Hassan Mohiuddin, Benjamin J Pedretti, Zachary P Smith, Jie Chen, Wojciech Matusik
- ICML 2024 Spotlight
- Readers: Everyone

View all 29 publications

Co-Authors

View all 74 co-authors