evaluation-task · Source-linked discovery

SciCode: A Research Coding Benchmark Curated by Scientists

SciCode tests the ability of language models to generate code to solve scientific research problems. It assesses models on 65 problems from mathematics, physics, chemistry, biology, and materials science.

Open interactive record →
OriginMinyang Tian, Luyu Gao, Shizhuo Dylan Zhang 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