EvoCodeBench: An Evolving Code Generation Benchmark with Domain-Specific Evaluations

• URL: http://arxiv.org/abs/2410.22821v1
• Date: Wed, 30 Oct 2024 08:57:59 GMT
• Title: EvoCodeBench: An Evolving Code Generation Benchmark with Domain-Specific Evaluations
• Authors: Jia Li, Ge Li, Xuanming Zhang, Yunfei Zhao, Yihong Dong, Zhi Jin, Binhua Li, Fei Huang, Yongbin Li,
• Abstract summary: Existing benchmarks have two limitations - data leakage and lack of domain-specific evaluation. EvoCodeBench will be dynamically updated every period (e.g., 6 months) to avoid data leakage. This paper releases the first version - EvoCodeBench-2403, containing 275 samples from 25 repositories.
• Score: 87.34429475432998
• License:
• Abstract: How to evaluate Large Language Models (LLMs) in code generation remains an open question. Existing benchmarks have two limitations - data leakage and lack of domain-specific evaluation. The former hurts the fairness of benchmarks, and the latter hinders practitioners from selecting superior LLMs for specific programming domains. To address these two limitations, we propose a new benchmark - EvoCodeBench, which has the following advances: (1) Evolving data. EvoCodeBench will be dynamically updated every period (e.g., 6 months) to avoid data leakage. This paper releases the first version - EvoCodeBench-2403, containing 275 samples from 25 repositories. (2) A domain taxonomy and domain labels. Based on the statistics of open-source communities, we design a programming domain taxonomy consisting of 10 popular domains. Based on the taxonomy, we annotate each sample in EvoCodeBench with a domain label. (3) Domain-specific evaluations. Besides the Pass@k, we compute the Domain-Specific Improvement (DSI) and define LLMs' comfort and strange domains. These evaluations help practitioners select superior LLMs in specific domains and discover the shortcomings of existing LLMs. We evaluate 8 popular LLMs (e.g., gpt-4, DeepSeek Coder) on EvoCodeBench and summarize some insights. EvoCodeBench reveals the actual abilities of these LLMs in real-world repositories. For example, the highest Pass@1 of gpt-4 on EvoCodeBench-2403 is only 20.74%. Besides, we evaluate LLMs in different domains and discover their comfort and strange domains. For example, gpt-4 performs best in most domains but falls behind others in the Internet domain. StarCoder 2-15B unexpectedly performs well in the Database domain and even outperforms 33B LLMs. EvoCodeBench has been released.

Related papers

• DOMAINEVAL: An Auto-Constructed Benchmark for Multi-Domain Code Generation [48.11754113512047]
  This study includes a code generation benchmark dataset DOMAINEVAL, encompassing six popular domains. Our pipeline works in a fully automated manner, enabling a push-bottom construction from code repositories into formatted subjects under study. The contributions of this study include a code generation benchmark dataset DOMAINEVAL, encompassing six popular domains, a fully automated pipeline for constructing code benchmarks, and an identification of the limitations of LLMs in code generation tasks based on their performance on DOMAINEVAL.
  arXiv  Detail & Related papers  (2024-08-23T16:33:58Z)
• DevEval: A Manually-Annotated Code Generation Benchmark Aligned with Real-World Code Repositories [83.5195424237358]
  Existing benchmarks are poorly aligned with real-world code repositories. We propose a new benchmark named DevEval, which has three advances. DevEval comprises 1,874 testing samples from 117 repositories, covering 10 popular domains.
  arXiv  Detail & Related papers  (2024-05-30T09:03:42Z)
• EvoCodeBench: An Evolving Code Generation Benchmark Aligned with Real-World Code Repositories [42.257427142180546]
  Existing benchmarks demonstrate poor alignment with real-world code repositories. EvoCodeBench is an evolving benchmark to avoid data leakage. Based on EvoCodeBench, we propose repository-level code generation and evaluate 10 popular Large Language Models.
  arXiv  Detail & Related papers  (2024-03-31T08:10:50Z)
• Top Leaderboard Ranking = Top Coding Proficiency, Always? EvoEval: Evolving Coding Benchmarks via LLM [13.324171480106715]
  EvoEval is a program synthesis benchmark suite created by evolving existing benchmarks into different targeted domains. Our study shows that compared to the high performance obtained on standard benchmarks like HumanEval, there is a significant drop in performance. We showcase various insights, including the brittleness of instruction-following models when encountering rewording or subtle changes.
  arXiv  Detail & Related papers  (2024-03-28T03:10:39Z)
• InfiBench: Evaluating the Question-Answering Capabilities of Code Large Language Models [56.723509505549536]
  InfiBench is the first large-scale freeform question-answering (QA) benchmark for code to our knowledge. It comprises 234 carefully selected high-quality Stack Overflow questions that span across 15 programming languages. We conduct a systematic evaluation for over 100 latest code LLMs on InfiBench, leading to a series of novel and insightful findings.
  arXiv  Detail & Related papers  (2024-03-11T02:06:30Z)
• On the Effectiveness of Large Language Models in Domain-Specific Code Generation [20.61882220430463]
  Large language models (LLMs) such as ChatGPT have shown remarkable capabilities in code generation. We investigate how to effectively incorporate API knowledge into the code generation process. We refer to these strategies as a new code generation approach called DomCoder.
  arXiv  Detail & Related papers  (2023-12-04T05:41:02Z)
• CMT in TREC-COVID Round 2: Mitigating the Generalization Gaps from Web to Special Domain Search [89.48123965553098]
  This paper presents a search system to alleviate the special domain adaption problem. The system utilizes the domain-adaptive pretraining and few-shot learning technologies to help neural rankers mitigate the domain discrepancy. Our system performs the best among the non-manual runs in Round 2 of the TREC-COVID task.
  arXiv  Detail & Related papers  (2020-11-03T09:10:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.