NormasTCU — A Brazilian Portuguese IR dataset and an evaluation of LLM-as-a-judge for relevance assessment
The results suggest that LLMs could effectively support scalable relevance assessment in specialized Portuguese corpora when evaluated using nDCG or MRR (rank-aware metrics), but they should be avoided when relying on precision or recall.
L. C. Fernandes, Marcus Vinicius Conceição de Castro, Leandro dos Santos Ribeiro et al.
· Information Processing &... · 0 citations