arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2505.04084cs.SEcs.AI
收藏

关于 Stack Overflow 上 OpenAI API 讨论的实证研究

An Empirical Study of OpenAI API Discussions on Stack Overflow

School of Artificial Intelligence and Computer Science, Nantong University(人工智能与计算机科学学院,南通大学);State Key Lab. for Novel Software Technology, Nanjing University(新型软件技术国家重点实验室,南京大学);Chang Chien College, Nantong University(长进学院,南通大学);Computer School, Beijing Information Science and Technology University(计算机学院,北京信息科技大学)

Xiang Chen, Jibin Wang, Chaoyang Gao, Xiaolin Ju, Zhanqi Cui

AI总结:

本文首次对 Stack Overflow 上 2,874 个 OpenAI API 相关讨论进行实证研究,通过人工分类和主题建模识别开发人员面临的主要挑战,并为开发者、LLM 供应商和研究者提出可操作建议。

摘要:

以 OpenAI 的 GPT 系列为代表的大语言模型(LLMs)的快速发展,已经对自然语言处理、软件开发、教育、医疗、金融和科学研究等多个领域产生了显著影响。然而,OpenAI API 带来了不同于传统 API 的独特挑战,例如提示工程的复杂性、基于 token 的成本管理、非确定性输出以及黑箱式运行方式。据我们所知,开发人员在使用 OpenAI API 时遇到的挑战在以往的实证研究中尚未被探索。为了填补这一空白,我们开展了首个全面的实证研究,分析了来自流行问答论坛 Stack Overflow 的 2,874 个与 OpenAI API 相关的讨论帖。我们首先考察了这些帖子的流行度和难度。在将其人工分类为九个与 OpenAI API 相关的类别之后,我们通过主题建模分析识别了每个类别所关联的具体挑战。基于我们的实证发现,我们最终为开发人员、LLM 供应商和研究人员提出了可操作的建议。

英文摘要:

The rapid advancement of large language models (LLMs), represented by OpenAI's GPT series, has significantly impacted various domains such as natural language processing, software development, education, healthcare, finance, and scientific research. However, OpenAI APIs introduce unique challenges that differ from traditional APIs, such as the complexities of prompt engineering, token-based cost management, non-deterministic outputs, and operation as black boxes. To the best of our knowledge, the challenges developers encounter when using OpenAI APIs have not been explored in previous empirical studies. To fill this gap, we conduct the first comprehensive empirical study by analyzing 2,874 OpenAI API-related discussions from the popular Q&A forum Stack Overflow. We first examine the popularity and difficulty of these posts. After manually categorizing them into nine OpenAI API-related categories, we identify specific challenges associated with each category through topic modeling analysis. Based on our empirical findings, we finally propose actionable implications for developers, LLM vendors, and researchers.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑