Evaluation design and benchmarks · MLD-0881
Pass@k
The probability or estimated fraction of tasks with at least one successful result among k sampled attempts.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Evaluate code generation with multiple candidates.
Original caveat
It does not mean every sample passed, and sampling assumptions matter.
Context and aliases
Level: Intermediate. Stages: Evaluation · Research.
Tags: ML · DL · LLM · SLM · Safety.
Aliases: pass at k.