I am a third-year PhD candidate in Computational Linguistics working on theInvestigaDiff project, where I evaluate and improve how well models capture subtle meaning differences across languages, especially in long documents. It is a pleasure to be supervised  by Prof. Dr. Rico Sennrich, Dr. Jannis Vamvas, and Prof. Dr. Sarah Ebling. Beyond my thesis work, I contribute to large-scale projects like Apertus, the Swiss open LLM, and the Last Translation Benchmark.

I hold a master's degree in Computational Linguistics and the interdisciplinary program Methods – Data – Society. For my master's thesis, I worked on unsupervised translation direction detection, which was recognized with the UZH semester award. I still enjoy methods work that finds simple solutions to well-defined problems, most recently scaling unsupervised word alignment to full documents.

Publications
Google Scholar

Apertus Team. 2026. Apertus 1.5: From Open Text Models to Open Multimodal Reasoners. Pre-print.  [model]

LTB Team. 2026. Last Translation Benchmark. Pre-print. [data] [code]

Michelle Wastl; Jannis Vamvas; Rico Sennrich (2026). Scaling Unsupervised Word Alignment to Documents via Structural Constraints. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 7419–7436, Budapest, Hungary. Association for Computational Linguistics. [code] [data] [demo]

Andrianos Michail; Stylianos Psychias; Michelle Wastl; Simon Clematide; Rico Sennrich; Juri Opitz ALEE: Any-Language Evaluation of Embeddings via English-Centric Minimal Pairs. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 7419–7436, Budapest, Hungary. Association for Computational Linguistics.

Michelle Wastl; Jannis Vamvas; Rico Sennrich (2026). SwissGov-RSD: A Human-annotated, Cross-lingual Benchmark for Token-level Recognition of Semantic Differences Between Related Documents. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics, San Diego, California, United States. Association for Computational Linguistics. [data] [code] ★ SAC highlight

Kevin Du; Clara Kümpel; Michelle Wastl; Alex Warstadt (2026). It’s Not What You Say, It’s How You Say It: Evaluating LLM Responses to Expressions of Belief. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics, San Diego, California, United States. Association for Computational Linguistics. [data] [code] 

Andrianos Michail; Stylianos Psychias; Michelle Wastl; Simon Clematide; Rico Sennrich; Juri Opitz (2026). ALEE: Any-Language Evaluation of Embeddings via English-Centric Minimal Pairs. Pre-print.[code]

Michelle Wastl; Jannis Vamvas; Rico Sennrich (2025).
Machine Translation Models are Zero-Shot Detectors of Translation Direction
In Findings of the Association for Computational Linguistics: ACL 2025. Vienna, Austria, pp. 1054–1074. [code] [demo]

Michelle Wastl; Jannis Vamvas; Rico Sennrich (2025).UZH at SemEval-2025 Task 3: Token-Level Self-Consistency for Hallucination Detection
In Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025). Vienna, Austria, pp. 257-270. [code] 

Michelle Wastl; Jannis Vamvas; Selena Calleri; Rico Sennrich (2025).
20min-XD: A Comparable Corpus of Swiss News Articles
In Proceedings of the 10th edition of the Swiss Text Analytics Conference. Winterthur, Switzerland. [code] [data] ★ best paper award

Selena Calleri; Michelle Wastl; Bojan Perić; Andreas Abegg (2024).
Interlex–A search engine to explore the interconnectedness of Swiss legal texts
In Proceedings of the 9th edition of the Swiss Text Analytics Conference. Chur, Switzerland, pp. 174-174.