Martin Potthast

ALiBi positional encoding can blind attention heads, drastically impairing token retrieval without impacting standard performance metrics.

Students learned that annotation disagreement is a valuable insight into the complexity of data interpretation, not merely a flaw in labeling.

Entity recognition models can effectively spot RAG-powered native ads, even when advertisers try to disguise them with different styles.

Algorithmic decisions about humans can now be audited for "Representation Fidelity" by checking if they align with self-reported descriptions, revealing potential biases and inaccuracies.