Summary of Evocodebench: An Evolving Code Generation Benchmark Aligned with Real-world Code Repositories, by Jia Li et al.
EvoCodeBench: An Evolving Code Generation Benchmark Aligned with Real-World Code Repositories
by Jia Li, Ge Li, Xuanming Zhang, Yihong Dong, Zhi Jin
First submitted to arxiv on: 31 Mar 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed benchmark, EvoCodeBench, addresses the limitations of existing benchmarks in evaluating Large Language Models (LLMs) in code generation. This medium-difficulty summary highlights that EvoCodeBench offers comprehensive annotations, robust evaluation metrics, and alignment with real-world repositories. The paper presents an automatic pipeline to update EvoCodeBench from the latest repositories, releasing its first version, EvoCodeBench-2403, containing 275 samples from 25 real-world repositories. This benchmark is used to evaluate 10 popular LLMs, revealing their coding abilities in real-world scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary EvoCodeBench is a new way to test how well Large Language Models (LLMs) can help with writing code. Right now, there’s no good way to measure if these models are really helping or not. This paper creates a special set of examples and rules called EvoCodeBench that tries to fix this problem. It makes sure the examples come from real-world coding projects, has detailed instructions for what the LLMs should do, and uses special metrics to see how well they’re doing. The researchers use this new benchmark to test 10 different LLMs and find out which ones are best at helping with code writing. |
Keywords
» Artificial intelligence » Alignment
