Benchmarking Benchmark Leakage in Large Language Models

TL;DR

This paper investigates the issue of benchmark dataset leakage in large language models, proposing a detection pipeline and transparency practices to improve evaluation fairness and transparency in the field.

Contribution

It introduces a scalable detection pipeline using Perplexity and N-gram accuracy and proposes the Benchmark Transparency Card for better documentation and transparency.

Findings

Substantial instances of benchmark leakage found in 31 LLMs

Leakage leads to potentially unfair model comparisons

Publicly released leaderboard and tools to facilitate future research

Abstract

Amid the expanding use of pre-training data, the phenomenon of benchmark dataset leakage has become increasingly prominent, exacerbated by opaque training processes and the often undisclosed inclusion of supervised data in contemporary Large Language Models (LLMs). This issue skews benchmark effectiveness and fosters potentially unfair comparisons, impeding the field's healthy development. To address this, we introduce a detection pipeline utilizing Perplexity and N-gram accuracy, two simple and scalable metrics that gauge a model's prediction precision on benchmark, to identify potential data leakages. By analyzing 31 LLMs under the context of mathematical reasoning, we reveal substantial instances of training even test set misuse, resulting in potentially unfair comparisons. These findings prompt us to offer several recommendations regarding model documentation, benchmark setup, and future evaluations. Notably, we propose the "Benchmark Transparency Card" to encourage clear documentation of benchmark utilization, promoting transparency and healthy developments of LLMs. we have made our leaderboard, pipeline implementation, and model predictions publicly available, fostering future research.