研究人员开发了SpecCoder，一个旨在通过利用中间形式化规范来增强代码大模型推理能力的新框架。与自然语言不同，这些可执行的规范为代码验证、调试和修复提供了机器可检查的约束。SpecCoder使用经过验证的程序、改变行为的变异体和精炼跟踪来训练模型，以生成对正确代码有效但拒绝错误代码的规范。该框架在HumanExec基准上进行了评估，显示出Qwen2.5-Coder等模型在规范质量、正确性和完整性方面有了显著的改进。 AI
影响 这项研究通过提高专用大模型的推理能力，可能带来更可靠的代码生成和验证工具。
排序理由 该集群包含一篇学术论文，详细介绍了用于改进代码大模型的新框架和基准。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2607.04232v1 Announce Type: cross Abstract: Unlike natural-language specifications, executable formal specifications provide machine-checkable constraints for verifying, debugging, and repairing code. However, writing such specifications is labor-intensive, and existing LLM…
Large language models can be adapted for low-resource programming languages through specialized training pipelines and benchmarks, achieving superior code completion performance compared to general-purpose models.
Unlike natural-language specifications, executable formal specifications provide machine-checkable constraints for verifying, debugging, and repairing code. However, writing such specifications is labor-intensive, and existing LLM-based methods mainly infer whole-program pre/post…