Replication materials for Dimension and support laws for monotone costly verification search
Year: 2026
Type: dataset
Abstract: This package contains the programs and retained synthetic outputs supporting a theoretical study of generation-paid, exact-verification search with monotone multivariate success probabilities. It includes rank and Pareto selectors, geometric and oracle checks, ordinal dependence experiments, and processing, mixed-margin and calibration certificates. The README gives the executable commands. Sampling outputs retain their parameter grids, seeds, environment versions and Monte Carlo standard errors. The experiments require Python and NumPy and do not call a language model or use an external dataset. Finite checks accompany, rather than replace, the manuscript's mathematical proofs. The archive contains 13 Python sources, 12 retained synthetic-result JSON files and a README. It does not contain the manuscript, author portrait, biography or submission forms.
Author: Fu, Junye
Institution: South China University of Technology
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221587384
Cites:
Cited by:
Related to: