DyePack: Provably Flagging Test Set Contamination in LLMs Using Backdoors

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Cheng, Yize, Wang, Wenxiao, Moayeri, Mazda, Feizi, Soheil
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909803710775296
author Cheng, Yize
Wang, Wenxiao
Moayeri, Mazda
Feizi, Soheil
author_facet Cheng, Yize
Wang, Wenxiao
Moayeri, Mazda
Feizi, Soheil
contents Open benchmarks are essential for evaluating and advancing large language models, offering reproducibility and transparency. However, their accessibility makes them likely targets of test set contamination. In this work, we introduce DyePack, a framework that leverages backdoor attacks to identify models that used benchmark test sets during training, without requiring access to the loss, logits, or any internal details of the model. Like how banks mix dye packs with their money to mark robbers, DyePack mixes backdoor samples with the test data to flag models that trained on it. We propose a principled design incorporating multiple backdoors with stochastic targets, enabling exact false positive rate (FPR) computation when flagging every model. This provably prevents false accusations while providing strong evidence for every detected case of contamination. We evaluate DyePack on five models across three datasets, covering both multiple-choice and open-ended generation tasks. For multiple-choice questions, it successfully detects all contaminated models with guaranteed FPRs as low as 0.000073% on MMLU-Pro and 0.000017% on Big-Bench-Hard using eight backdoors. For open-ended generation tasks, it generalizes well and identifies all contaminated models on Alpaca with a guaranteed false positive rate of just 0.127% using six backdoors.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23001
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DyePack: Provably Flagging Test Set Contamination in LLMs Using Backdoors
Cheng, Yize
Wang, Wenxiao
Moayeri, Mazda
Feizi, Soheil
Computation and Language
Open benchmarks are essential for evaluating and advancing large language models, offering reproducibility and transparency. However, their accessibility makes them likely targets of test set contamination. In this work, we introduce DyePack, a framework that leverages backdoor attacks to identify models that used benchmark test sets during training, without requiring access to the loss, logits, or any internal details of the model. Like how banks mix dye packs with their money to mark robbers, DyePack mixes backdoor samples with the test data to flag models that trained on it. We propose a principled design incorporating multiple backdoors with stochastic targets, enabling exact false positive rate (FPR) computation when flagging every model. This provably prevents false accusations while providing strong evidence for every detected case of contamination. We evaluate DyePack on five models across three datasets, covering both multiple-choice and open-ended generation tasks. For multiple-choice questions, it successfully detects all contaminated models with guaranteed FPRs as low as 0.000073% on MMLU-Pro and 0.000017% on Big-Bench-Hard using eight backdoors. For open-ended generation tasks, it generalizes well and identifies all contaminated models on Alpaca with a guaranteed false positive rate of just 0.127% using six backdoors.
title DyePack: Provably Flagging Test Set Contamination in LLMs Using Backdoors
topic Computation and Language
url https://arxiv.org/abs/2505.23001