按年被提到
20232023 · 6 次2026
软件工程与 AI 编程›代码生成与程序合成·arXiv:2305.01210 · cs.SE

Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation

Jiawei Liu, Chunqiu Steven Xia, Yuyao Wang, Lingming Zhang · 2023
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 3 人首次 2023-07-11 · 最近 2023-08-26
3不同的人
2活跃月份
5讨论串
502被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Program synthesis has been long studied with recent approaches focused on directly using the power of Large Language Models (LLMs) to generate code. Programming benchmarks, with curated synthesis problems and test-cases, are used to measure the performance of various LLMs on code synthesis. However, these test-cases can be limited in both quantity and quality for fully assessing the functional correctness of the generated code. Such limitation in the existing benchmarks begs the following question: In the era of LLMs, is the code generated really correct? To answer this, we propose EvalPlus -- a code synthesis evaluation framework to rigorously benchmark the functional correctness of LLM-synthesized code. EvalPlus augments a given evaluation dataset with large amounts of test-cases newly produced by an automatic test input generator, powered by both LLM- and mutation-based strategies. While EvalPlus is general, we extend the test-cases of the popular HumanEval benchmark by 80x to build HumanEval+. Our extensive evaluation across 26 popular LLMs (e.g., GPT-4 and ChatGPT) demonstrates that HumanEval+ is able to catch significant amounts of previously undetected wrong code synthesized by LLMs, reducing the pass@k by up-to 19.3-28.9%. We also surprisingly found that test insufficiency can lead to mis-ranking. For example, both WizardCoder-CodeLlama and Phind-CodeLlama now outperform ChatGPT on HumanEval+, while none of them could on HumanEval. Our work not only indicates that prior popular code synthesis evaluation results do not accurately reflect the true performance of LLMs for code synthesis, but also opens up a new direction to improve such programming benchmarks through automated testing. We have open-sourced our tools, enhanced datasets as well as all LLM-generated code at https://github.com/evalplus/evalplus to facilitate and accelerate future LLM-for-code research.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论lhl · 2023-07

“Since I've been on a AI code-helper kick recently. According to the post, Claude 2 now 71.2%, a significant upgrade from 1.3 (56.0%). It isn't specified whether this is pass@1 or pass@10. For comparison: * GPT-4 claims 85.4 on HumanEval, in a recent paper https://arxiv.org/pdf/2303.11366.pdf GPT-4 was tested at 80.1 pass@1 and 91 pass@1 using their Refl…”

评论lhl · 2023-07

“While that's what the Technical Report (https://arxiv.org/pdf/2303.08774v3.pdf) says, but GPT-4 out in the wild's (reproducible) performance appears to be much higher now. Testing from 3/15 (presumably on the 0314 model) seems to be at 85.36% (https://twitter.com/amanrsanger/status/1635751764577361921). And the linked paper from my post(https://doi.org/10…”

评论og_kalu · 2023-07

“GPT-4 out in the wild's (reproducible) performance appears to be much higher than 67. Testing from 3/15 (presumably on the 0314 model) seems to be at 85.36% (https://twitter.com/amanrsanger/status/1635751764577361921). And the linked paper from my post(https://doi.org/10.48550/arXiv.2305.01210) got a pass@1 of 88.4 from GPT-4 recently (May? June?).”

Claude 2 Internal API Client and CLI原评论 ↗
评论lhl · 2023-07

“I have not read the paper yet (in my backlog, here's the paper: https://arxiv.org/pdf/2307.09009.pdf), but it's important note that the paper is entitled "How Is ChatGPT’s Behavior Changing over Time?" not that it's necessarily "getting worse." Here's a more nuanced (not an AI clout chasing account) discussion by Arvind Narayanan (Princeton CS prof) discu…”

GPT-4 is getting worse over time, not better原评论 ↗
评论EvgeniyZh · 2023-08

“Note that current GPT-4 pass@1 for HumanEval is closer to 90% than to 67% reported in GPT-4 technical report, as reported, e.g., in [1] [1] https://arxiv.org/abs/2305.01210”

Code Llama, an AI tool for coding原评论 ↗
评论EvgeniyZh · 2023-08

“I have a several arguments why contamination is probably not the main reason of performance difference. When we worked on StarCoder, people ran gpt-4 on MultiPL-E, which doesn't have canonical solutions in the internet, and the performance was higher that what you would expect from official numbers Official contamination analysis shows only minor drop i…”

Show HN: Beating GPT-4 on HumanEval with a fine-tuned CodeLlama-34B原评论 ↗