研究人员推出了JAMER，这是一个新的数据集和基准测试，旨在评估AI模型在专业游戏引擎中的项目级代码生成能力。JAMER利用游戏开发大赛的数据，专注于Godot引擎，包含8,133个已验证的项目。该基准测试使用编译通过率、结构完整性得分和行为对齐得分等指标，评估了驱动式生成和代码补全等任务。初步评估显示，随着项目复杂度的增加，AI模型的性能显著下降，突显了架构设计是关键瓶颈。 AI
影响 凸显了当前AI代码生成在复杂项目级任务（尤其是在游戏开发领域）中的局限性。
排序理由 该集群描述了一个用于AI代码生成的新数据集和基准测试，该测试发表在arXiv论文中。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.19830v1 Announce Type: cross Abstract: Current AI-driven game development has made substantial progress in asset generation, gameplay design, and web-based game coding, yet project-level code engineering on professional game engines remains largely unexplored due to th…
Current AI-driven game development has made substantial progress in asset generation, gameplay design, and web-based game coding, yet project-level code engineering on professional game engines remains largely unexplored due to the absence of large-scale datasets and deterministi…
Game development frameworks and benchmarks were created using data from game jam competitions to evaluate code generation and project-level programming capabilities.