Scholar
Jonas Gehring
Google Scholar ID: jOwTwm4AAAAJ
Facebook AI Research
Citations & Impact
Citations
7,257
H-index
15
i10-index
15
Publications
20
Co-authors
11
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers including but not limited to:
- - 'RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning' (ICML 2025)
- - 'What Makes Large Language Models Reason in (Multi-Turn) Code Generation?' (ICLR 2024)
- - 'The Larger the Better? Improved LLM Code-Generation via Budget Reallocation' (COLM 2024)
- - 'Code Llama: Open Foundation Models for Code'
- - 'Leveraging Demonstrations with Latent Space Priors' (TMLR 03/2023)
- - 'Hierarchical Skills for Efficient Exploration' (NeurIPS 2021)
- - 'Forward Modeling for Partial Observation Strategy Games - A StarCraft Defogger' (NeurIPS 2018)
- - 'High-Level Strategy Selection under Partial Observability in StarCraft: Brood War' (RLPO@NeurIPS 2018)
- - 'STARDATA: A StarCraft AI Research Dataset'
Research Experience
- Worked as a research assistant at the Interactive Systems Lab at Karlsruhe Institute of Technology in 2013, focusing on speech recognition and deep learning.
Education
- Since 2019, pursuing a PhD at ETH Zürich under Prof. Andreas Krause, and collaborating with Nicolas Usunier and Gabriel Synnaeve at Facebook AI Research.
- In 2012, completed a Master's degree at Karlsruhe Institute of Technology, focusing on Machine Learning, Cognitive Systems, and Theoretical Computer Science. Master's thesis involved applying deep learning techniques to feature preprocessing in automatic speech recognition, supervised by Alex Waibel and Florian Metze, and supported by InterACT, conducted at Carnegie Mellon University.
- In 2009, completed undergraduate studies in Computer Science at the University of Freiburg. Bachelor's thesis titled 'Intelligente Objekterkennung für ein lernfähiges Carrerabahn-System', supervised by Prof. Martin Riedmiller.
Background
- Research interests: Neural Networks, Machine Learning, Cognitive Systems, Algorithms
- Programming interests: Parallelization, Optimization, Languages
- Other interests: Reading, Guitar (Acoustic, Electric), Drums, Biking, Snowboarding
Miscellany
- Preferred Linux distribution: Arch
