Certified reference models for AI evals: deterministically broken in one documented way each, so a benchmark's detection rate is measurable against ground truth.
-
Updated
Oct 9, 2026 - Python
#
defect-injection
Here are 2 public repositories matching this topic...
Reproducible benchmarks, Halmos formal verification proofs, safety-corpus defect detection, and cost/throughput ledgers.
security smart-contracts benchmarks formal-verification ai-safety runtime-verification halmos defect-injection
-
Updated
Jun 11, 2026 - Solidity
Add this topic to your repo
To associate your repository with the defect-injection topic, visit your repo's landing page and select "manage topics."