Benchmark Almanac

NICT

We track 3 benchmarks that NICT made or helped make.

0 in use 0 with public code 0 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
Asian Language Treebank Models translate parallel sentences across Asian languages. The llm-jp-eval harness uses the parallel corpus for machine translation. Translation Not in cards
AudioMOS Challenge 2025 Models predict listener ratings for generated audio. Three tracks assess music quality and prompt alignment, general audio aesthetics, and speech quality across sampling rates. Audio understanding and voice Mar 2025 Not in cards
Japanese-English Bilingual Corpus of Wikipedia’s Kyoto Articles Models translate between Japanese and English using parallel Wikipedia passages about Kyoto. Translation Oct 2010 Not in cards