Papers by Klaudia Thellmann
Quantifying the Impact of Translation Errors on Multilingual LLM Evaluation (2026.acl-long)
Copied to clipboard
| Challenge: | Machine-translated benchmarks are widely used to assess the multilingual capabilities of large language models (LLMs), yet translation errors in these benchmarks remain underexplored. |
| Approach: | They show how well machine-translated benchmarks match human span annotations on translations . they also show how strongly translation errors explain accuracy drops on translated benchmarks - a gap that is not addressed yet . |
| Outcome: | The proposed model matches human-level translations with human-language annotations on translations, but translation errors are associated with accuracy drops even after controlling for English correctness and source-side anomalies. |
Investigating Multilingual Instruction-Tuning: Do Polyglot Models Demand for Multilingual Instructions? (2024.emnlp-main)
Copied to clipboard
Alexander Weber, Klaudia Thellmann, Jan Ebert, Nicolas Flores-Herr, Jens Lehmann, Michael Fromm, Mehdi Ali
| Challenge: | a study of multilingual pre-trained LLMs on parallel instruction-tuning benchmarks shows that instruction-following models can be used across languages by up to 9.9%. |
| Approach: | They conduct an extensive study of the performance of multilingual pre-trained LLMs instruction-tuned on parallel instruction-uning datasets. |
| Outcome: | The proposed model improves cross-lingual instruction following capabilities by 9.9% on a large and mid-sized LLM on parallel instruction-tuning datasets. |
Tokenizer Choice For LLM Training: Negligible or Crucial? (2024.findings-naacl)
Copied to clipboard
Mehdi Ali, Michael Fromm, Klaudia Thellmann, Richard Rutmann, Max Lübbering, Johannes Leveling, Katrin Klug, Jan Ebert, Niclas Doll, Jasper Buschhoff, Charvi Jain, Alexander Weber, Lena Jurkschat, Hammam Abdelwahab, Chelsea John, Pedro Ortiz Suarez, Malte Ostendorff, Samuel Weinbach, Rafet Sifa, Stefan Kesselheim, Nicolas Flores-Herr
| Challenge: | Recent success of large language models has been driven by curating the training dataset composition, scaling of model architectures and advancements in pretraining objectives, leaving tokenizer influence as a blind spot. |
| Approach: | They conduct a comprehensive study on the influence of tokenizer choice on LLM downstream performance by training 24 mono- and multilingual LLMs at a 2.6B parameter scale. |
| Outcome: | The proposed model can significantly impact the model's downstream performance and training costs. |