SQALE: A Large Text-to-SQL Corpus Grounded in Real Schemas
Wolff, C., Gomm, D., & Hulsebos, M. SQaLe: A large text-to-SQL corpus grounded in real schemas. In EurIPS 2025 Workshop: AI for Tabular Data.
CV
Education
- Ph.D. in Computer Science, Centrum Wiskunde & Informatica & University of Amsterdam (2025–Present)
- Research focus: Insight Retrieval from Structured Data
- Supervisors: Madelon Hulsebos, Maarten de Rijke
- M.Sc. in Cognitive Science, Osnabrück University, Germany (2021–2025)
- GPA: 1.0 (equivalent to A+)
- Focus: Machine Learning, Computational Neuroscience, Ethics of AI
- Thesis: Emergence of language in situated environments (Grade: 1.0)
- B.Sc. in Information Systems, Osnabrück University, Germany (2018–2021)
- GPA: 2.1
- Thesis: Simulation environment for analyzing the spread of SARS-CoV-2 using ML agents (Grade: 1.0)
Work Experience
- PhD Student, Centrum Wiskunde & Informatica & University of Amsterdam (Mar 2025 – Present)
- Insight extraction from structured data
- Research pipeline design and model development
- Junior Researcher, German Research Center for AI (DFKI), Osnabrück (Mar 2022 – Mar 2025)
- Led work package in Green-AI Hub with 5 contributors
- Projects: In-context image classification, RAG systems, LLM evaluation
- Promoted from Research Assistant based on prior work
- Research Assistant, Elia Bruni’s Lab, Osnabrück University (Apr 2023 – Present)
- Co-authored multiple papers (including one as first author)
- Built 3D RL experiment framework using MuJoCo
- Research Assistant, Distributed Systems Lab, Osnabrück University (Jun 2021 – Mar 2023)
- First-author peer-reviewed paper
- Focus: Distributed systems and scientific data analysis
- Freelance AI Developer (Remote, Jun 2021 – Mar 2023)
- Projects: RAG, synthetic data for embeddings, NLP tools
- Soft skills: Client relations, project coordination
Skills
- Programming: Python (PyTorch, TensorFlow, Pandas), JavaScript/TypeScript, Tailwind CSS, SQL
- ML/NLP: LLMs, RAG, RL, SNNs, Vector Databases
- Tools: Git, Bash/Zsh, Linux, LaTeX (Overleaf, R Markdown)
Publications
Are We Asking the Right Questions? On Ambiguity in Natural Language Queries for Tabular Data Analysis
Gomm, D., Wolff, C., & Hulsebos, M. (2025). Are We Asking the Right Questions? On Ambiguity in Natural Language Queries for Tabular Data Analysis. AI for Tabular Data Workshop at EuRIPS 2025.
MinervAI: Using Generative AI to Assist, Not Replace Humans in Peer Review
Hüsing, I., Petersen, T., Weiher, M.-A., Wolff, C., & Musslick, S. (2025). MinervAI: Using Generative AI to Assist, Not Replace Humans in Peer Review. AI in Science (AIS) 2025.
How well do LLMs reason over tabular data, really?
Wolff, C., & Hulsebos, M. (2025). "How well do LLMs reason over tabular data, really?" 4th Table Representation Workshop - ACL 2025.
PictSure: Pretraining Embeddings Matters for In-Context Learning Image Classifiers
Schiesser, L., Wolff, C., Haas, S., & Pukrop, S. (2025). "PictSure: Pretraining Embeddings Matters for In-Context Learning Image Classifiers." arXiv preprint arXiv:2506.14842.
Bidirectional Emergent Language in Situated Environments
Wolff, C., Mayer, J., Bruni, E., & Ohmer, X. (2024). "Bidirectional Emergent Language in Situated Environments." arXiv preprint arXiv:2408.14649.
GRASP: A Novel Benchmark for Evaluating Language Grounding and Situated Physics Understanding in Multimodal Language Models
Jassim, S., Holubar, M., Richter, A., Wolff, C., Ohmer, X., & Bruni, E. (2024). "GRASP: A Novel Benchmark for Evaluating Language Grounding and Situated Physics Understanding in Multimodal Language Models." Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24, 6297-6305.
A Review of Nine Physics Engines for Reinforcement Learning Research
Kaup, M., Wolff, C., Hwang, H., Mayer, J., & Bruni, E. (2024). "A Review of Nine Physics Engines for Reinforcement Learning Research." arXiv preprint arXiv:2407.08590.
A New Approach on Estimating Germany’s Mobile Broadband Coverage based on Crowdsourced Data
Wolff, C., Tessmer, A., & Aschenbruck, N. (2023). "A New Approach on Estimating Germany's Mobile Broadband Coverage based on Crowdsourced Data." Mobile Communication-Technologies and Applications; 27th ITG-Symposium, 31-36.
Talks
How well do LLMs reason over tabular data?
Online Talk at AlphaXiv, Amsterdam, Netherlands
Can We Create Green AI? - Podcast Interview
Podcast Interview at Kaleidoscience: Conversations on Cognitive Science, Osnabrück, Germany
Teaching
Service and Leadership
- Master’s application selection committee member (2023–2025)
- Led research project groups (20+ students) in EBIMAS
- Supervised 4 bachelor theses
- Led work package at DFKI with interdisciplinary collaboration