evaluation-task · Source-linked discovery

USACO: USA Computing Olympiad

Evaluates language model performance on difficult Olympiad programming problems across four difficulty levels.

Open interactive record →
OriginQuan Shi, Michael Tang, Karthik Narasimhan et al.TopicsGeneral capabilityStatusimported

Can support

Not independently assessed by FronteraEval yet.

Cannot support by itself

No inference beyond the upstream source should be made until the protocol is reviewed.

Original sources