How does the pass@k metric for code generation models vary across BigCodeBench tasks requiring multi-library P

Abstract

Abstract: Repeated sampling with a verifier is the standard way to allocate test-time compute for code generation, with pass@\$K\$ as the canonical metric. Yet the standard policy class draws \$K\$ independent samples from a single answer distribution, so attempts often collapse onto near-duplicate reasoning paths and waste the budget on redundant rollouts. This failure is costly in competitive programming, where many problems admit multiple distinct algorithmic strategies and pass@\$K\$ requires only one correct attempt. We propose Coordinated Pass@\$K\$ Policy Optimization (CPPO), which turns pass@\$K\$ generat

Research Question

How does the pass@k metric for code generation models vary across BigCodeBench tasks requiring multi-library Python data science workflows compared to single-library tasks?

Verification Level

Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.

Quality Tier

Descriptive public triage only; this tier does not alter current publication or DOI behavior.

Quality Dimensions

Automated triage signals derived from public fields; not human peer review or independent validation.

Correction Record

Public corrections are additive records. Current status does not claim the synthesis is error-free.

Provenance