KoLasSimpleQA is a new multilingual simple factuality benchmark with general and language-specific domains, showing LLMs are consistently weaker on language-specific knowledge across nine languages.
The head entity and the relation of the triple are used to generate the question, and the tail entity is the answer
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluating Large Language Model with Knowledge Oriented Language Specific Simple Question Answering
KoLasSimpleQA is a new multilingual simple factuality benchmark with general and language-specific domains, showing LLMs are consistently weaker on language-specific knowledge across nine languages.