# ConStat: Performance-Based Contamination Detection in Large Language Models

arXiv: https://arxiv.org/abs/2405.16281 · HTML: https://mlsanity.com/paper/2405.16281/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 53 references that are in the map.
- First posted on arXiv 25 May 2024.

## Builds on (most cited first, 30 of 38)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [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)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [TIES-Merging: Resolving Interference When Merging Models](https://mlsanity.com/paper/2306.01708/) (2023)
- [Crowdsourcing Multiple Choice Science Questions](https://mlsanity.com/paper/1707.06209/) (2017)
- [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)
- [Detecting Pretraining Data from Large Language Models](https://mlsanity.com/paper/2310.16789/) (2023)
- [NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmark](https://mlsanity.com/paper/2310.18018/) (2023)
- [Membership Inference Attacks against Language Models via Neighbourhood Comparison](https://mlsanity.com/paper/2305.18462/) (2023)
- [Proving Test Set Contamination in Black Box Language Models](https://mlsanity.com/paper/2310.17623/) (2023)
- [Rethinking Benchmark and Contamination for Language Models with Rephrased Samples](https://mlsanity.com/paper/2311.04850/) (2023)
- [Don't Make Your LLM an Evaluation Benchmark Cheater](https://mlsanity.com/paper/2311.01964/) (2023)
- [Arcee's MergeKit: A Toolkit for Merging Large Language Models](https://mlsanity.com/paper/2403.13257/) (2024)
- [A Careful Examination of Large Language Model Performance on Grade School Arithmetic](https://mlsanity.com/paper/2405.00332/) (2024)
- [Time Travel in LLMs: Tracing Data Contamination in Large Language Models](https://mlsanity.com/paper/2308.08493/) (2023)
- [Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks](https://mlsanity.com/paper/2203.03929/) (2022)
- [Investigating Data Contamination in Modern Benchmarks for Large Language Models](https://mlsanity.com/paper/2311.09783/) (2023)

## Built on by (most cited first, 30 of 33)

- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluation](https://mlsanity.com/paper/2502.17521/) (2025)
- [Time Travel is Cheating: Going Live with DeepFund for Real-Time Fund Investment Benchmarking](https://mlsanity.com/paper/2505.11065/) (2025)
- [Eureka: Evaluating and Understanding Large Foundation Models](https://mlsanity.com/paper/2409.10566/) (2024)
- [MathConstruct: Challenging LLM Reasoning with Constructive Proofs](https://mlsanity.com/paper/2502.10197/) (2025)
- [How Contaminated Is Your Benchmark? Quantifying Dataset Leakage in Large Language Models with Kernel Divergence](https://mlsanity.com/paper/2502.00678/) (2025)
- [AdaCoder: An Adaptive Planning and Multi-Agent Framework for Function-Level Code Generation](https://mlsanity.com/paper/2504.04220/) (2025)
- [Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation](https://mlsanity.com/paper/2602.09305/) (2026)
- [Detecting Data Contamination from Reinforcement Learning Post-training for Large Language Models](https://mlsanity.com/paper/2510.09259/) (2025)
- [Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence](https://mlsanity.com/paper/2608.31075/) (2026)
- [A Statistical Framework for Auditing Behavioral Dependence and Induced Bias in LLM Judges](https://mlsanity.com/paper/2604.07650/) (2026)
- [Streamlining evidence based clinical recommendations with large language models](https://mlsanity.com/paper/2505.10282/) (2025)
- [Lean Meets Theoretical Computer Science: Scalable Synthesis of Theorem Proving Challenges in Formal-Informal Pairs](https://mlsanity.com/paper/2508.15878/) (2025)
- [LLM-as-a-Judge is Bad, Based on AI Attempting the Exam Qualifying for the Member of the Polish National Board of Appeal](https://mlsanity.com/paper/2511.04205/) (2025)
- [TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models](https://mlsanity.com/paper/2605.26161/) (2026)
- [Temporal Leakage in LLM Backtesting: Measurement, Validation, and Adjusted Scores](https://mlsanity.com/paper/2608.02985/) (2026)
- [AdEval: Alignment-based Dynamic Evaluation to Mitigate Data Contamination in Large Language Models](https://mlsanity.com/paper/2501.13983/) (2025)
- [Provable Training Data Identification for Large Language Models](https://mlsanity.com/paper/2510.09717/) (2025)
- [Obscuring Data Contamination Through Translation: Evidence from Arabic Corpora](https://mlsanity.com/paper/2601.14994/) (2026)
- [Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications](https://mlsanity.com/paper/2605.26133/) (2026)
- [The Reliability Gap in Benchmark Auditing: Distribution Shift and Scale as Failure Modes of Contamination Detection](https://mlsanity.com/paper/2606.03305/) (2026)
- [Same Question, Different Answers: Evaluating LLM Reliability Beyond Accuracy](https://mlsanity.com/paper/2607.22554/) (2026)
- [Contamination Inflates Scores but Rarely Reorders Large Language Model Leaderboards](https://mlsanity.com/paper/2609.02899/) (2026)
- [RN-F: A Novel Approach for Mitigating Contaminated Data in Large Language Models](https://mlsanity.com/paper/2505.13249/) (2025)
- [Silicon Bureaucracy and AI Test-Oriented Education: Contamination Sensitivity and Score Confidence in LLM Benchmarks](https://mlsanity.com/paper/2603.21636/) (2026)
- [On Emotion-Sensitive Decision Making of Small Language Model Agents](https://mlsanity.com/paper/2604.06562/) (2026)
- [SPENCE: A Syntactic Probe for Detecting Contamination in NL2SQL Benchmarks](https://mlsanity.com/paper/2604.17771/) (2026)
- [Data Contamination in Neural Hieroglyphic Translation: A Reproducibility Study](https://mlsanity.com/paper/2605.07453/) (2026)
- [Physics-R1: An Audited Olympiad Corpus and Released Verifiers for Visual Physics Reasoning](https://mlsanity.com/paper/2605.14040/) (2026)
- [Correcting test set contamination by spiking the training data](https://mlsanity.com/paper/2605.24818/) (2026)

## Abstract (arXiv)

Public benchmarks play an essential role in the evaluation of large language models. However, data contamination can lead to inflated performance, rendering them unreliable for model comparison. It is therefore crucial to detect contamination and estimate its impact on measured performance. Unfortunately, existing detection methods can be easily evaded and fail to quantify contamination. To overcome these limitations, we propose a novel definition of contamination as artificially inflated and non-generalizing benchmark performance instead of the inclusion of benchmark samples in the training data. This perspective enables us to detect any model with inflated performance, i.e., performance that does not generalize to rephrased samples, synthetic samples from the same distribution, or different benchmarks for the same task. Based on this insight, we develop ConStat, a statistical method that reliably detects and quantifies contamination by comparing performance between a primary and reference benchmark relative to a set of reference models. We demonstrate the effectiveness of ConStat in an extensive evaluation of diverse model architectures, benchmarks, and contamination scenarios and find high levels of contamination in multiple popular models including Mistral, Llama, Yi, and the top-3 Open LLM Leaderboard models.