CodeCriticBench: A Holistic Code Critique Benchmark for Large Language Models

TL;DR

CodeCriticBench is a comprehensive benchmark designed to evaluate the critique abilities of large language models across diverse code-related tasks and multiple evaluation dimensions.

Contribution

It introduces a holistic evaluation framework for LLM critique capacity, covering multiple code tasks and detailed assessment protocols, addressing limitations of prior benchmarks.

Findings

Existing benchmarks lack comprehensive code critique evaluation.

CodeCriticBench effectively evaluates critique abilities across tasks and dimensions.

Experimental results demonstrate the benchmark's effectiveness.

Abstract

The critique capacity of Large Language Models (LLMs) is essential for reasoning abilities, which can provide necessary suggestions (e.g., detailed analysis and constructive feedback). Therefore, how to evaluate the critique capacity of LLMs has drawn great attention and several critique benchmarks have been proposed. However, existing critique benchmarks usually have the following limitations: (1). Focusing on diverse reasoning tasks in general domains and insufficient evaluation on code tasks (e.g., only covering code generation task), where the difficulty of queries is relatively easy (e.g., the code queries of CriticBench are from Humaneval and MBPP). (2). Lacking comprehensive evaluation from different dimensions. To address these limitations, we introduce a holistic code critique benchmark for LLMs called CodeCriticBench. Specifically, our CodeCriticBench includes two mainstream code tasks (i.e., code generation and code QA) with different difficulties. Besides, the evaluation protocols include basic critique evaluation and advanced critique evaluation for different characteristics, where fine-grained evaluation checklists are well-designed for advanced settings. Finally, we conduct extensive experimental results of existing LLMs, which show the effectiveness of CodeCriticBench.