Seacrowd A Multilingual Multimodal Data Hub And Benchmark Suite For Southeast Asian Languages 2024 06 14
Source
cohere.com
Author
Cohere
Date
Context
SEACrowd brings Southeast Asian language resources into a standardized collection spanning text, images, and audio. The paper's research summary, listed on Cohere's research pages, describes resources covering nearly 1,000 languages and a benchmarkA standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.Full definition → evaluating 36 indigenous languages across 13 tasks.
Those numbers describe different kinds of coverage. The larger collection supplies research material; the narrower benchmark defines where the reported model evalA repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.Full definition → took place. The summary supports using SEACrowd to broaden language research and testing, without implying that models have been assessed in every language represented in the collection.
Terms in this piece · Glossary
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
postCohere releases open-weights translation model North Small Translate
articleThe Data Provenance Initiative A Large Scale Audit Of Dataset Licensing And Attribution In Ai 2023 10 25
articleSeataubench Progressively Adapting Tool Agent User Evaluation Into Low Resource Southeast Asian Languages 2026 06 27Cohere editorial sitemap
articleTiny Aya Bridging Scale And Multilingual Depth 2026 02 17Cohere
articleInclude Evaluating Multilingual Language Understanding With Regional Knowledge 2024 11 29Cohere