Glossary · Optimization and failure dynamics
Generator coupling
Dependence between the generator's failure modes and the evaluator's, measured as Pr[E errs | G errs] versus Pr[E errs], or by comparing false-accept rates on the generator's failures with those on other sources' failures.
Defined in
Handbook · Validity under pressureGoodhart pressure and self-confirming loopsReward hacking needs an exploit; self-confirmation needs only a shared blind spot. Tell them apart and measure the coupling behind both.Related terms
- Self-confirming loop
- Generator and evaluator share correlated blind spots, so selection or training amplifies common errors. It needs no adversary, only correlated bias.
- Differential misclassification
- Evaluator error rates that differ between compared systems, for example because one was optimized against the evaluator. One shared calibration then biases corrections and comparisons; audit error per system or at least per generator family.
Where it comes up
- Evaluators inside training loopsHandbook · Regime
- The validation profile and validity statementHandbook · Validity under pressure
- Self-improvement and evaluator co-evolutionHandbook · Agents and self-improvement
- Anti-patterns and failure vocabularyHandbook · Practice
- Notation and formula cardHandbook · Reference