lightbulb现实痛点编程助手与多轮代码支持机器学习开发者缺乏高效工具和流程来处理模型错误,影响模型部署。CHI '23Faulty or Ready? Handling Failures in Deep-Learning Computer Vision Models until Deployment: A Study of Practices, Challenges, and Needs
lightbulb现实痛点编程助手与多轮代码支持开发者往往过度信任AI工具生成的代码,导致严重的安全风险。CHI '25Exploring the Impact of Intervention Methods on Developers’ Security Behavior in a Manipulated ChatGPT Study
lightbulb现实痛点编程助手与多轮代码支持程序员在Notebook中因全局变量影响难以追踪变量和避免命名冲突。CHI '25Pagebreaks: Multi-Cell Scopes in Computational Notebooks
lightbulb现实痛点编程助手与多轮代码支持程序员在使用代码工具时需频繁明确上下文和请求,引发效率低下。CHI '25Need Help? Designing Proactive AI Assistants for Programming
lightbulb现实痛点编程助手与多轮代码支持程序员使用低阶调试工具时,信息过载且难以直观把握数据结构。CHI '25The Shapes of Abstraction in Data Structure Diagrams
lightbulb现实痛点编程助手与多轮代码支持非技术用户很难高效构建个性化的机器学习流程。CHI '25InstructPipe: Generating Visual Blocks Pipelines with Human Instructions and LLMs