Alina Fastowski

According to our database, Alina Fastowski authored at least 6 papers between 2023 and 2026.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of five.

2026

Analysing the Safety Pitfalls of Steering Vectors.

Yuxiao Li, Alina Fastowski, Efstratios Zaradoukas, Bardh Prenkaj, Gjergji Kasneci

CoRR, March, 2026

2025

Injecting Falsehoods: Adversarial Man-in-the-Middle Attacks Undermining Factual Recall in LLMs.

Alina Fastowski, Bardh Prenkaj, Yuxiao Li, Gjergji Kasneci

CoRR, November, 2025

Attention Mechanisms Don't Learn Additive Models: Rethinking Feature Importance for Transformers.

Tobias Leemann, Alina Fastowski, Felix Pfeiffer, Gjergji Kasneci

Trans. Mach. Learn. Res., 2025

From Confidence to Collapse in LLM Factual Robustness.

Alina Fastowski, Bardh Prenkaj, Gjergji Kasneci

Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2025, 2025

2024

Understanding Knowledge Drift in LLMs Through Misinformation.

Alina Fastowski, Gjergji Kasneci

Proceedings of the Discovering Drift Phenomena in Evolving Landscapes, 2024

2023

Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity.

Katharina Hämmerl, Alina Fastowski, Jindrich Libovický, Alexander Fraser

Proceedings of the Findings of the Association for Computational Linguistics: ACL 2023, 2023