Abstract:Recently, many large language models (LLMs) have been proposed, showing advanced proficiency in code generation. Meanwhile, many efforts have been dedicated to evaluating LLMs on code generation benchmarks such as HumanEval. Although being very helpful for comparing different LLMs, existing evaluation focuses on a sim-ple code generation scenario (i.e., function-level or statement-level code generation), which mainly asks LLMs to generate one single code unit (e.g., a function or a statement) for the given natural language description. Such evaluation focuses on generating independent and often small-scale code units, thus leaving it unclear how LLMs perform in real-world software development scenarios. To fill this knowledge gap, we make the first attempt to evaluate LLMs in a more challenging code generation scenario, i.e., class-level code generation. Compared with existing code generation benchmarks, it better reflects real-world software development scenarios due to it comprising broader contextual dependencies and multiple, interdependent units of code. We first manually construct the first class-level code generation benchmark ClassEval of 100 class-level Python code generation tasks with approximately 500 person-hours. Based on the new benchmark ClassEval, we then perform the first study of 11 state-of-the-art LLMs on class-level code generation. Based on our results, we find that all LLMs perform much worse on class-level code generation compared to the method-level. While GPT models still dominate other LLMs on class-level code generation, the performance rankings of other models on method-level code generation no longer holds for class-level code generation. Besides, most models (except GPT models) perform better when generating the class method by method; and they have the limited ability of generating dependent code. Based on our findings, we call for software engineering (SE) researchers' expertise to build more LLM benchmarks based on practical and complicated software development scenarios.

The Emergence of Large Language Models in Static Analysis: A First Look through Micro-Benchmarks

An Empirical Study of Large Language Models for Type and Call Graph Analysis

Large Language Models for Code Analysis: Do LLMs Really Do Their Job?

LMs: Understanding Code Syntax and Semantics for Code Analysis

Beyond Static Tools: Evaluating Large Language Models for Cryptographic Misuse Detection

An analysis of large language models: their impact and potential applications

Large Language Models: A Survey

Understanding the Effectiveness of Large Language Models in Detecting Security Vulnerabilities

Large Linguistic Models: Analyzing theoretical linguistic abilities of LLMs

LLM4DS: Evaluating Large Language Models for Data Science Code Generation

Evaluating Large Language Models in Detecting Test Smells

Benchmarking Large Language Models for Log Analysis, Security, and Interpretation

An Evaluation of Large Language Models in Bioinformatics Research

A Preliminary Study on Using Large Language Models in Software Pentesting

A Survey of Large Language Models for Code: Evolution, Benchmarking, and Future Trends

Large Language Models for Secure Code Assessment: A Multi-Language Empirical Study

Evaluating Large Language Models in Class-Level Code Generation

Large Language Models as Code Executors: An Exploratory Study

Large Language Models on Graphs: A Comprehensive Survey

On the Evaluation of Large Language Models in Unit Test Generation

Inadequacies of Large Language Model Benchmarks in the Era of Generative Artificial Intelligence