2024/03/31 by Yiqing Xie, Xie, Yiqing, Alex Xie +9 · 7 citations
Computer Science · #Computation and Language (cs.CL) #Embedded Systems Design Techniques #FOS: Computer and information sciences #Parallel Computing and Optimization Techniques #Software Engineering (cs.SE) #Software Testing and Debugging Techniques
paper · pdf · doi:10.48550/arxiv.2404.00566
openalex publication_date 2024/03/31 · openalex created_date 2024/04/04 · openalex updated_date 2026/07/28
To adequately test modern code generation systems, evaluation benchmarks must execute and test the code generated by the system. However, these execution and testing requirements have largely limited benchmarks to settings where code is easily executable or has human-written tests. To facilitate evaluation of code generation systems across diverse scenarios, we present CodeBenchGen, a framework to create scalable execution-based benchmarks from naturally occurring code sources. Specifically, we leverage a large language model (LLM) to sandbox arbitrary pieces of code into evaluation examples, including test cases for execution-based evaluation. We illustrate the usefulness of our framework by creating a dataset, Exec-CSN, which includes 1,931 examples involving 293 libraries converted from code in 367 GitHub repositories taken from the Code- SearchNet dataset. To demonstrate the solvability of examples in Exec-CSN, we present a human study demonstrating that 81.3% of the examples can be solved by humans and 61% are rated as "requires effort to solve". We conduct code generation experiments on open-source and proprietary models and analyze the performance of both humans and models. We provide code and data at: https://github.com/yiqingxyq/CodeBenchGen.