My current research projects are closely connected to my recent publications. Each project page collects the paper, code, datasets, demos, and follow-up material where available.

LREC 2026 · Benchmark QA

Diagnosing translated benchmarks

Automated quality assurance for translated benchmark items.

Paper · Code · Data · Slides ACL 2026 · Translation Quality

Translation errors in multilingual LLM evaluation

Quantifying how translation errors affect multilingual model evaluation.

Paper · Code · Data · Demo OpenGPT-X · 2022–2025 · NAACL · EMNLP · EACL

OpenGPT-X: Teuken-7B & multilingual evaluation

Open European LLMs and the tokenizer, instruction-tuning, and evaluation work behind them.

Papers · Models · Leaderboard · Code