Skip to content

Topic

Agent stopping rules

Policies that decide when an iterative AI agent loop should stop, such as fixed iteration budgets, score thresholds, plateau detection and learned or Bayesian stopping criteria.

Current clusters

build1 publisher

A noisy judge drags an adaptive agent stop rule below a fixed six-step budget

CDV's adaptive stop rule truly reached its quality bar in 71% of runs under judge noise of 0.10, against 93% for a fixed six-step budget. Requiring two passing scores in a row lifts that to 97% for 1.4 extra steps, so a one-line change beats both the Bayesian policy and the budget.

Publishers:dev.to

Reality

Evidence58
Adoption
Insufficient
Hype gap+8
Incentives30
Confidence55