test_grpo_advantages_center_per_group (test_advantages.AdvantageTest.test_grpo_advantages_center_per_group) ... ok
test_rloo_excludes_self (test_advantages.AdvantageTest.test_rloo_excludes_self) ... ok
test_bootstrap_is_seeded (test_metrics.MetricsTest.test_bootstrap_is_seeded) ... ok
test_paired_ci_rejects_shape_mismatch (test_metrics.MetricsTest.test_paired_ci_rejects_shape_mismatch) ... ok
test_pass_at_k (test_metrics.MetricsTest.test_pass_at_k) ... ok
test_error_recovery_replays_exactly (test_replay.ReplayTest.test_error_recovery_replays_exactly) ... ok
test_unknown_action_has_clear_error (test_replay.ReplayTest.test_unknown_action_has_clear_error) ... ok
test_strict_requires_canonical_answer (test_verifier.VerifierTest.test_strict_requires_canonical_answer) ... ok
test_weak_verifier_exposes_hack (test_verifier.VerifierTest.test_weak_verifier_exposes_hack) ... ok

----------------------------------------------------------------------
Ran 9 tests in 1.345s

OK
