Work

Preregistration: trivial-policy and model scores on first-wave benchmark packages, before and after independent review

Year: 2026

Type: article

Abstract: Preregistration of a small evaluation study. Hypotheses, the fixed list of packages, versions, trivial policies and models, the analysis plan, the exclusion rules and the stopping rule, written before the main data were collected. A pilot on one package is disclosed in the document.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.23424272

Author: Arab, Pooria

Language: English

Topic: Ethics and Social Impacts of AI

Subfield: Safety Research

Field: Social Sciences

Domain: Social Sciences

Open Access status: green

OpenAlex ID: W7222744624

FWCI:

Cites:

Cited by:

Related to: