Simulating Training Data Leakage in Multiple-Choice Benchmarks for LLM Evaluation
Hidayat, Naila Shafirni Al Kautsar, Muhammad Dehan Wicaksono, Alfan Farizki Koto, Fajri
Hidayat, Naila Shafirni
Al Kautsar, Muhammad Dehan
Wicaksono, Alfan Farizki
Koto, Fajri
Supervisor
Department
Natural Language Processing
Embargo End Date
Type
Conference proceeding
Date
License
Language
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
The performance of large language models (LLMs) continues to improve, as reflected in rising scores on standard benchmarks. However, the lack of transparency around training data raises concerns about potential overlap with evaluation sets and the fairness of reported results. Although prior work has proposed methods for detecting data leakage, these approaches primarily focus on identifying outliers and have not been evaluated under controlled simulated leakage conditions. In this work, we compare existing leakage detection techniques, namely permutation and n-gram-based methods, under a continual pretraining setup that simulates real-world leakage scenarios, and additionally explore a lightweight method we call semi-half question. We further introduce two efficient extensions, permutation-R and permutation-Q. While semi-half offers a low-cost alternative, our analysis shows that the n-gram method consistently achieves the highest F1-Score, performing competitively with permutation-Q. We also refine these techniques to support instance-level detection and reduce computational overhead. Leveraging the best-performing method, we create cleaned versions of MMLU and HellaSwag, and re-evaluate several LLMs. Our findings present a practical path toward more reliable and transparent evaluations, and we recommend contamination checks as a standard practice before releasing benchmark results.
Citation
N.S. Hidayat, M.D. Al Kautsar, A.F. Wicaksono, F. Koto, "Simulating Training Data Leakage in Multiple-Choice Benchmarks for LLM Evaluation," 2025, pp. 21-39.
Source
Proceedings of the 5th Workshop on Evaluation and Comparison of NLP Systems
Conference
Proceedings of the 5th Workshop on Evaluation and Comparison of NLP Systems
Keywords
Subjects
Source
Proceedings of the 5th Workshop on Evaluation and Comparison of NLP Systems
Publisher
Association for Computational Linguistics