# Addressing Data Leakage in HumanEval Using Combinatorial Test Design

arXiv: https://arxiv.org/abs/2412.01526 · HTML: https://mlsanity.com/paper/2412.01526/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 23 references that are in the map.
- First posted on arXiv 2 Dec 2024.

## Builds on (most cited first, 12 of 12)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation](https://mlsanity.com/paper/2101.11718/) (2021)
- [Large Language Models for Software Engineering: A Systematic Literature Review](https://mlsanity.com/paper/2308.10620/) (2023)
- [NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmark](https://mlsanity.com/paper/2310.18018/) (2023)
- [Don't Make Your LLM an Evaluation Benchmark Cheater](https://mlsanity.com/paper/2311.01964/) (2023)
- [Large Language Models for Software Engineering: Survey and Open Problems](https://mlsanity.com/paper/2310.03533/) (2023)
- [Software Testing with Large Language Models: Survey, Landscape, and Vision](https://mlsanity.com/paper/2307.07221/) (2023)
- [Task Contamination: Language Models May Not Be Few-Shot Anymore](https://mlsanity.com/paper/2312.16337/) (2023)
- [Benchmark Self-Evolving: A Multi-Agent Framework for Dynamic LLM Evaluation](https://mlsanity.com/paper/2402.11443/) (2024)
- [Dynamic Evaluation of Large Language Models by Meta Probing Agents](https://mlsanity.com/paper/2402.14865/) (2024)
- [A Software Engineering Perspective on Testing Large Language Models: Research, Practice, Tools and Benchmarks](https://mlsanity.com/paper/2406.08216/) (2024)

## Built on by (most cited first, 1 of 1)

- [A Metamorphic Testing Approach to Diagnosing Memorization in LLM-Based Program Repair](https://mlsanity.com/paper/2604.21579/) (2026)

## Abstract (arXiv)

The use of large language models (LLMs) is widespread across many domains, including Software Engineering, where they have been used to automate tasks such as program generation and test classification. As LLM-based methods continue to evolve, it is important that we define clear and robust methods that fairly evaluate performance. Benchmarks are a common approach to assess LLMs with respect to their ability to solve problem-specific tasks as well as assess different versions of an LLM to solve tasks over time. For example, the HumanEval benchmark is composed of 164 hand-crafted tasks and has become an important tool in assessing LLM-based program generation. However, a major barrier to a fair evaluation of LLMs using benchmarks like HumanEval is data contamination resulting from data leakage of benchmark tasks and solutions into the training data set. This barrier is compounded by the black-box nature of LLM training data which makes it difficult to even know if data leakage has occurred. To address the data leakage problem, we propose a new benchmark construction method where a benchmark is composed of template tasks that can be instantiated into new concrete tasks using combinatorial test design. Concrete tasks for the same template task must be different enough that data leakage has minimal impact and similar enough that the tasks are interchangeable with respect to performance evaluation. To assess our benchmark construction method, we propose HumanEval_T, an alternative benchmark to HumanEval that was constructed using template tasks and combinatorial test design.