研究人员开发了RuC，一个用于生成硬件描述语言（HDL）代码补全基准的新框架。该系统由语法驱动且语言无关，能够对寄存器传输级（RTL）开发中的大型语言模型（LLMs）进行受控评估。RuC根据HDL语法屏蔽代码区域，并提示模型重新生成它们，从而能够评估从简单赋值到整个逻辑块的能力。一项使用RuC在Tiny Tapeout和RISC-V核心的SystemVerilog基准上进行的研究表明，补全性能受模型类型、屏蔽区域结构和提示策略的影响，其中Fill-in-the-Middle（FIM）取得了最佳结果。 AI
影响 为在RTL开发中评估LLMs提供了一种更细粒度、更受控的方法，有可能提高模型在硬件设计任务中的性能。
排序理由 学术论文，介绍了一种用于硬件描述语言中LLMs的新基准生成框架。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2604.27780v1 Announce Type: cross Abstract: Large Language Models (LLMs) have rapidly improved in performance across code-related tasks, making their integration into Register Transfer Level (RTL) development increasingly attractive. Mimicking the behavior of inline code as…
Large Language Models (LLMs) have rapidly improved in performance across code-related tasks, making their integration into Register Transfer Level (RTL) development increasingly attractive. Mimicking the behavior of inline code assistants, many benchmarks evaluate LLMs' capabilit…
Large Language Models (LLMs) have rapidly improved in performance across code-related tasks, making their integration into Register Transfer Level (RTL) development increasingly attractive. Mimicking the behavior of inline code assistants, many benchmarks evaluate LLMs' capabilit…