🧑‍🔬 Matthias Bethge

🏛️ Affiliations

🎓University of TübingenTübingen · Germany2009–2026 · 195 paperscurrent🎓University of LübeckLübeck · Germany2014–2026 · 153 paperscurrent🎓Tübingen AI CenterTübingen · Germany2019–2026 · 40 paperscurrent🔬Max Planck Institute for Biological CyberneticsTübingen · Germany1900–2022 · 166 papers🔬Max Planck SocietyMunich · Germany1900–2019 · 116 papers🏛️Bernstein Center for Computational Neuroscience TübingenTübingen · Germany2010–2024 · 65 papers🎓Medizinische Hochschule HannoverHanover · Germany2011–2016 · 31 papers🏥Charité - Universitätsmedizin BerlinBerlin · Germany2008–2015 · 17 papers🎓Baylor College of MedicineHouston · United States2018–2021 · 13 papers🎓University of BremenBremen · Germany1999–2019 · 10 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 10 more institutions on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility (2025 · 36 citations)
   Progress in mathematical reasoning for language models is often judged on small benchmarks under inconsistent evaluation setups.
2. CiteME: Can Language Models Accurately Cite Scientific Claims? (2024 · 25 citations)
   The paper asks whether language models can work as research assistants for citation attribution: given a passage from a paper with a masked reference, can the model name the cited work? The authors first audited existing citation-prediction datasets.
3. Wu's Method can Boost Symbolic AI to Rival Silver Medalists and AlphaGeometry to Outperform Gold Medalists at IMO Geometry (2024 · 7 citations)
   Automated proving of olympiad geometry is often treated as a milestone for machine reasoning. AlphaGeometry is a neuro-symbolic system trained on 100 million synthetic proofs. It solved 25 of the 30 problems in the IMO-AG-30 benchmark, while the Wu's method baseline reported alongside it managed only ten. The authors…

All works

11 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

1. Image Style Transfer Using Convolutional Neural Networks ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetLeon A. Gatys, Alexander S. Ecker, Matthias Bethge2016 · 5,947 citations
2. Shortcut Learning in Deep Neural Networks ↗arXiv:2004.07780 · no summary yetRobert Geirhos et al.2020 · 2,366 citations
3. Controlling Perceptual Factors in Neural Style Transfer ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetLeon A. Gatys et al.2017 · 495 citations
4. Excessive Invariance Causes Adversarial Vulnerability ↗arXiv (Cornell University) · no summary yetJörn-Henrik Jacobsen et al.2018 · 65 citations
5. A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to ReproducibilityarXiv:2504.07086Andreas Hochlehnert, Hardik Bhatnagar, Vishaal Udandarao, et al.2025 · 36 citations
6. CiteME: Can Language Models Accurately Cite Scientific Claims?Conference on Neural Information Processing Systems (NeurIPS)Ori Press, Andreas Hochlehnert, Ameya Prabhu, et al.2024 · 25 citations
7. Wu's Method can Boost Symbolic AI to Rival Silver Medalists and AlphaGeometry to Outperform Gold Medalists at IMO GeometryarXiv:2404.06405Shiven Sinha, Ameya Prabhu, Ponnurangam Kumaraguru, et al.2024 · 7 citations
8. A Practitioner's Guide to Continual Multimodal Pretraining ↗arXiv (Cornell University) · no summary yetKarsten Roth et al.2024 · 4 citations
9. A Practitioner's Guide to Real-World Continual Multimodal Pretraining ↗neural information processing systems · no summary yetVishaal Udandarao et al.2024 · 4 citations
10. Disentanglement and Generalization Under Correlation Shifts ↗arXiv (Cornell University) · no summary yetChristina M. Funke et al.2021 · 1 citation
11. Unintended cue learning: Lessons for deep learning from experimental psychology ↗Journal of Vision · no summary yetRobert Geirhos et al.2020 · 1 citation

🤝 Co-authors here

6 co-authors with a page, the most papers together first.

• Ameya Prabhu ×3
• Andreas Hochlehnert ×2
• Vishaal Udandarao ×2
• Ofir Press
• Ponnurangam Kumaraguru
• Samuel Albanie

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings ×2
• Conference on Neural Information Processing Systems (NeurIPS): NeurIPS 2024
• Journal of Vision
• neural information processing systems

🔗 Cited by

Papers of our library cite their works 446 times (of the 149,439 papers whose bibliographies we read); 72 citing papers with a page here →.

Updated 3 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/matthias-bethge/