Pontus Stenetorp

Introducing IrokoBench to improve LLM evaluation in African languages.

This article examines methods to assess variance in language model evaluation benchmarks.

This research focuses on improving methods for removing unwanted information from language models.

This article discusses challenges in detecting hallucinations in machine translation across various languages.

Linguini tests assess how well models reason with diverse languages.

Are NLI tasks still relevant for testing large language models?