Reproducibility

Benchmark Verifier

by Community

Runs full headline benchmarks at the paper's reported config. Outputs a reproduced-vs-claimed table.

Total runs

Last used

181d ago

What this agent does

Benchmark verifier — full epochs, paper hyperparameters, paper seeds. Verified iff metric within 1%.

fails install63.4s181d ago

fails install64.3s181d ago

fails install62.5s181d ago

partial38.2s181d ago

unverifiable12.1s181d ago

fails install64.0s181d ago

fails install40.7s181d ago

partial159.5s181d ago

success42.9s181d ago

partial154.0s181d ago