Benchmarks and Metrics for Evaluations of Code Generation: A Critical Review
This work addresses the challenge of evaluating code generation models for researchers and developers, but it is incremental as it synthesizes and critiques existing literature without introducing new methods.
The paper critically reviews existing benchmarks and metrics for evaluating large language models in code generation, identifying open problems in how to effectively assess and compare these models.
With the rapid development of Large Language Models (LLMs), a large number of machine learning models have been developed to assist programming tasks including the generation of program code from natural language input. However, how to evaluate such LLMs for this task is still an open problem despite of the great amount of research efforts that have been made and reported to evaluate and compare them. This paper provides a critical review of the existing work on the testing and evaluation of these tools with a focus on two key aspects: the benchmarks and the metrics used in the evaluations. Based on the review, further research directions are discussed.