Skip to main content
Search

# Pareto Dominance

1 related articles

Subscribe to updates
  1. arXiv

    Multiple verifiers stop collapsing into one score: POLPO trains on Pareto rankings and raises full-pass rates in code and instruction following

    In code generation and instruction following, one response often has to pass several checks, and the usual practice aggregates those outcomes into a scalar reward before policy optimization. POLPO keeps only the comparisons the verification itself supports: one response is better when it passes every check the other passes and at least one more, while responses with conflicting outcomes stay incomparable. The objective maximizes the total probability of all complete rankings consistent with these partial orders, and on controlled program synthesis, MBPP, APPS and multi-constraint instruction following it yields higher rates of satisfying every requirement than all-pass or pass-rate reward training.

You’re all caught up