TAG: TESTING (2)
Scaffolding vs. Intelligence: Why AI Evaluation Harnesses Fake 70-Point Leaps
Why AI evaluation harness architectures create massive score spreads on benchmarks like ARC-AGI-3, and how to establish strict benchmark provenance.
When 100% Test Coverage Lies: Contract Faults and the Least-Derived Evidence Principle
A deep dive into why green unit tests can pass while specifications mandate contradictions. Learn the least-derived evidence principle for reliable software contracts.