Lukas Schott

Researcher, Aleph Alpha

Joined

February 2018

Names

Lukas Schott

Emails

****@bethgelab.org (Confirmed)

****@de.bosch.com (Confirmed)

****@bosch.com (Confirmed)

****@aleph-alpha.com (Confirmed)

Personal Links

Google Scholar

DBLP

Career & Education History

Researcher

Aleph Alpha (aleph-alpha.com)

2025 – Present

Researcher

Research, Bosch (bosch.com)

2021 – 2025

PhD student

Eberhard-Karls-Universität Tübingen (uni-tuebingen.de)

2017 – 2021

Advisors, Relations & Conflicts

PhD Advisee

Wieland Brendel

2017 – 2021

PhD Advisor

Mathias Bethge

2017 – 2021

Expertise

LLM Pretraining data

2025 – Present

knowledge distillation

2021 – 2024

Publications

Foundation Model-Based Data Selection for Dense Prediction Tasks
Niclas Popp, Dan Zhang, Jan Hendrik Metzen, Matthias Hein, Lukas Schott
- ICLR 2025 FM-Wild Workshop
- Readers: Everyone
Attention Is All You Need For Mixture-of-Depths Routing
Advait Gadhikar, Souptik Kumar Majumdar, Niclas Popp, Piyapat Saranrittichai, Martin Rapp, Lukas Schott
- SCOPE - ICLR 2025 Poster
- Readers: Everyone
Mind the Gap Between Synthetic and Real: Probing Transfer Capabilities of Stable Diffusion Images
Leonhard Hennicke, Christian Medeiros Adriano, Holger Giese, Jan Mathias Koehler, Lukas Schott
- Compression Workshop @ NeurIPS 2024
- Readers: Everyone
Investigating the Role of Fine-Tuning in Addressing the Gap Between Synthetic and Real Data in Generative Foundation Models
Leonhard Hennicke, Christian Medeiros Adriano, Holger Giese, Lukas Schott, Jan Mathias Koehler
- FITML 2024 Poster
- Readers: Everyone
Attention Is All You Need For Mixture-of-Depths Routing
Advait Gadhikar, Souptik Kumar Majumdar, Niclas Popp, Piyapat Saranrittichai, Martin Rapp, Lukas Schott
- Submitted to ICLR 2025
- Readers: Everyone
Mind the Gap Between Synthetic and Real: Utilizing Transfer Learning to Probe the Boundaries of Stable Diffusion Generated Data
Leonhard Hennicke, Christian Medeiros Adriano, Holger Giese, Jan Mathias Köhler, Lukas Schott
- CoRR 2024
- Readers: Everyone
Attention Is All You Need For Mixture-of-Depths Routing
Advait Gadhikar, Souptik Kumar Majumdar, Niclas Popp, Piyapat Saranrittichai, Martin Rapp, Lukas Schott
- CoRR 2024
- Readers: Everyone
Object-Focused Data Selection for Dense Prediction Tasks
Niclas Popp, Dan Zhang, Jan Hendrik Metzen, Matthias Hein, Lukas Schott
- CoRR 2024
- Readers: Everyone
Analytical Uncertainty-Based Loss Weighting in Multi-Task Learning
Lukas Kirchdorfer, Cathrin Elich, Simon Kutsche, Heiner Stuckenschmidt, Lukas Schott, Jan Mathias Köhler
- CoRR 2024
- Readers: Everyone
Understanding Neural Coding on Latent Manifolds by Sharing Features and Dividing Ensembles
Martin Bjerke, Lukas Schott, Kristopher T. Jensen, Claudia Battistin, David A. Klindt, Benjamin Adric Dunn
- ICLR 2023
- Readers: Everyone

View all 31 publications

Co-Authors

View all 51 co-authors