Search papers, labs, and topics across Lattice.
3
0
2
11
LLMs excel at recognizing true cognates but falter dramatically with false friends, exposing a fundamental flaw in cross-lingual semantic reasoning.
A structured taxonomy reveals the complexities of Arabic grammatical errors, paving the way for more effective language model evaluations.
Instruction-tuned LLMs can nearly match supervised baselines on complex Arabic morphosyntactic tagging and dependency parsing, but only with careful prompt engineering and retrieval-based in-context learning.
