Search papers, labs, and topics across Lattice.
Leibniz University Hannover
2
0
3
VEXMLM boosts performance on Amharic and Tigrinya by over 20 points in question answering, showcasing the power of tailored vocabulary for low-resource languages.
LLMs may ace English, but LLM Probe reveals surprising performance disparities in low-resource languages, with sequence-to-sequence models unexpectedly leading in morphosyntax.