Sample record. This work is fictional demonstration content. Results, people and institutions are placeholders; references in full-text articles are real publications.
Abstract
Benchmark scores can be inflated when test material reaches the training data or when evaluation choices quietly favour a model. We review 140 papers and incident reports and organise the mechanisms of evaluation leakage into a taxonomy with four top-level classes and eleven sub-classes. For each class we summarise how it can be detected, how often it has been reported, and what practices reduce its likelihood.
Keywords benchmark contamination · evaluation · leakage · survey
Full text
CTM-PP-2026-0004-v2.pdf
Declarations
- Author contributions
- Author contributions are listed in the PDF.
- Funding
- Sample record – no funding to declare.
- Competing interests
- The authors declare no competing interests.
- Data and code availability
- Sample article. All results, names and affiliations in this demonstration article are synthetic placeholders.
Versions
Version 2 · 14 October 2026Latest
Added four incident reports and one sub-class.
10.xxxx/ctm.pp.2026.0004.v2
Version 1 · 31 August 2026
First version.
10.xxxx/ctm.pp.2026.0004.v1
How to cite
Haddad, N., Novak, I., & Mensah, K. (2026). A Taxonomy of Evaluation Leakage in Language-Model Benchmarks [Preprint, version 2]. CONTINUUM Preprints. doi:10.xxxx/ctm.pp.2026.0004
© 2026 The authors. Published by AEX CONTINUUM under the CC BY 4.0 licence.