# Data Contamination Can Cross Language Barriers

arXiv: https://arxiv.org/abs/2406.13236 · HTML: https://mlsanity.com/paper/2406.13236/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 19 Jun 2024.

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms](https://mlsanity.com/paper/1905.13319/) (2019)
- [Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus](https://mlsanity.com/paper/2104.08758/) (2021)
- [Scalable Extraction of Training Data from (Production) Language Models](https://mlsanity.com/paper/2311.17035/) (2023)
- [Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERT](https://mlsanity.com/paper/1904.09077/) (2019)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [Time Travel in LLMs: Tracing Data Contamination in Large Language Models](https://mlsanity.com/paper/2308.08493/) (2023)
- [Platypus: Quick, Cheap, and Powerful Refinement of LLMs](https://mlsanity.com/paper/2308.07317/) (2023)
- [Investigating Data Contamination for Pre-training Language Models](https://mlsanity.com/paper/2401.06059/) (2024)
- [Quantifying Contamination in Evaluating Code Generation Capabilities of Language Models](https://mlsanity.com/paper/2403.04811/) (2024)
- [Estimating Contamination via Perplexity: Quantifying Memorisation in Language Model Evaluation](https://mlsanity.com/paper/2309.10677/) (2023)
- [MindMerger: Efficient Boosting LLM Reasoning in non-English Languages](https://mlsanity.com/paper/2405.17386/) (2024)

## Built on by (most cited first, 26 of 26)

- [Preference Leakage: A Contamination Problem in LLM-as-a-judge](https://mlsanity.com/paper/2502.01534/) (2025)
- [On Memorization of Large Language Models in Logical Reasoning](https://mlsanity.com/paper/2410.23123/) (2024)
- [SoK: Membership Inference Attacks on LLMs are Rushing Nowhere (and How to Fix It)](https://mlsanity.com/paper/2406.17975/) (2024)
- [A Survey on Data Contamination for Large Language Models](https://mlsanity.com/paper/2502.14425/) (2025)
- [The Bitter Lesson Learned from 2,000+ Multilingual Benchmarks](https://mlsanity.com/paper/2504.15521/) (2025)
- [Towards Data Contamination Detection for Modern Large Language Models: Limitations, Inconsistencies, and Oracle Challenges](https://mlsanity.com/paper/2409.09927/) (2024)
- [Forget What You Know about LLMs Evaluations -- LLMs are Like a Chameleon](https://mlsanity.com/paper/2502.07445/) (2025)
- [CapBencher: Give Your LLM Benchmark a Built-in Alarm for Test-Set Overfitting](https://mlsanity.com/paper/2505.18102/) (2025)
- [OWL: Probing Cross-Lingual Recall of Memorized Texts via World Literature](https://mlsanity.com/paper/2505.22945/) (2025)
- [When Flores Bloomz Wrong: Cross-Direction Contamination in Machine Translation Evaluation](https://mlsanity.com/paper/2601.20858/) (2026)
- [Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems](https://mlsanity.com/paper/2410.23472/) (2024)
- [LogitTrace: Detecting Benchmark Contamination via Layerwise Logit Trajectories](https://mlsanity.com/paper/2509.20909/) (2025)
- [LiveCLKTBench: Towards Reliable Evaluation of Cross-Lingual Knowledge Transfer in Multilingual LLMs](https://mlsanity.com/paper/2511.14774/) (2025)
- [Quantifying the Effect of Test Set Contamination on Generative Evaluations](https://mlsanity.com/paper/2601.04301/) (2026)
- [How Open Must Language Models be to Enable Reliable Scientific Inference?](https://mlsanity.com/paper/2603.26539/) (2026)
- [To See is Not to Learn: Protecting Multimodal Data from Unauthorized Fine-Tuning of Large Vision-Language Model](https://mlsanity.com/paper/2605.14291/) (2026)
- [AdEval: Alignment-based Dynamic Evaluation to Mitigate Data Contamination in Large Language Models](https://mlsanity.com/paper/2501.13983/) (2025)
- [RePCS: Diagnosing Data Memorization in LLM-Powered Retrieval-Augmented Generation](https://mlsanity.com/paper/2506.15513/) (2025)
- [Obscuring Data Contamination Through Translation: Evidence from Arabic Corpora](https://mlsanity.com/paper/2601.14994/) (2026)
- [Cultivar: A Contrastive and Locale-Oriented Translation Benchmark for Investigating Contamination and Localisation Robustness](https://mlsanity.com/paper/2608.09766/) (2026)
- [Toward Robust Multilingual Adaptation of LLMs for Low-Resource Languages](https://mlsanity.com/paper/2510.14466/) (2025)
- [When AI Navigates the Fog of War](https://mlsanity.com/paper/2603.16642/) (2026)
- [Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity](https://mlsanity.com/paper/2605.06327/) (2026)
- [Uncertainty-based Debiasing and Unlearning for Decontamination](https://mlsanity.com/paper/2606.23313/) (2026)
- [MultiSynt/MT: Trillion-Token Multi-Parallel Pre-Training Data Translated Across 36 Languages](https://mlsanity.com/paper/2607.00890/) (2026)
- [Detecting Pretraining Data in Large Language Models from a Free-Energy Perspective](https://mlsanity.com/paper/2609.21888/) (2026)

## Abstract (arXiv)

The opacity in developing large language models (LLMs) is raising growing concerns about the potential contamination of public benchmarks in the pre-training data. Existing contamination detection methods are typically based on the text overlap between training and evaluation data, which can be too superficial to reflect deeper forms of contamination. In this paper, we first present a cross-lingual form of contamination that inflates LLMs' performance while evading current detection methods, deliberately injected by overfitting LLMs on the translated versions of benchmark test sets. Then, we propose generalization-based approaches to unmask such deeply concealed contamination. Specifically, we examine the LLM's performance change after modifying the original benchmark by replacing the false answer choices with correct ones from other questions. Contaminated models can hardly generalize to such easier situations, where the false choices can be not even wrong, as all choices are correct in their memorization. Experimental results demonstrate that cross-lingual contamination can easily fool existing detection methods, but not ours. In addition, we discuss the potential utilization of cross-lingual contamination in interpreting LLMs' working mechanisms and in post-training LLMs for enhanced multilingual capabilities. The code and dataset we use can be obtained from urlhttps://github.com/ShangDataLab/Deep-Contam.