Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Conference on Neural Information Processing Systems (NeurIPS)

Venue URL:

The Percentage of Empirical Papers Documenting Each Reproducibility Variable

Venue
Reproducibility Score based on Gundersen et al. (2025). See Methods for details.
Documentation Score is the average score over the seven reproducibility variables for empirical research papers. See Methods for details.
Percentage of papers that are empirical research vs theoretical research.
Percentage of empirical research papers with at least one author from Industry.
Website
NeurIPS 2025 5286 0.69 4.89 95.63% 35.23%
NeurIPS 2024 4035 0.69 4.65 93.63% 39.25%
NeurIPS 2023 3218 0.6 4.05 92.42% 40.99%
NeurIPS 2022 2671 0.64 3.94 90.94% 43.85%
NeurIPS 2021 2334 0.58 3.66 90.66% 46.22%
NeurIPS 2020 1898 0.56 3.38 89.88% 43.26%
NeurIPS 2019 1428 0.52 3.31 89.64% 43.05%
NeurIPS 2018 1009 0.45 3.13 90.19% 40.99%
NeurIPS 2017 679 0.39 2.95 90.13% 35.62%
NeurIPS 2016 569 0.35 2.88 86.99% 29.09%
NeurIPS 2015 403 0.35 2.94 88.34% 25.0%
NeurIPS 2014 411 0.3 2.78 89.54% 22.83%