跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20935 · v1

Razavi-Bench: Tracking AI Progress in Analog Circuit Reasoning

首次提交 2026-11-16 19:45
版本更新 2026-11-16 19:45
雷达观测 2026-11-17 17:11
arXiv ↗PDF ↗

摘要

We propose Razavi-Bench, a benchmark built from the 50 questions in Razavi's two Analog Design Experiments With AI articles and graded by GPT-6 Luna against reference answers and rubrics, and use it to track how fast AI models improve at analog circuit analysis. Across 73 multimodal large language models released between May 2024 and November 2026, the top score rose from 28.83% to 98.33% in 30 months, mostly before the reference answers were published. The leading models now answer almost every textbook question correctly; the hardest questions show a circuit that resembles a textbook one but works differently, or ask about behavior the obvious analysis misses. Within two and a half years, a set of questions that exposed the limits of AI models has become one they answer almost perfectly.

作者备注:6 pages, 3 figures, 2 tables. Project page: https://razavi-bench.tokenzhang.com ; code and data: https://github.com/Arcadia-1/razavi-bench

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 19:45 · 最新观测版本

Razavi-Bench: Tracking AI Progress in Analog Circuit Reasoning

We propose Razavi-Bench, a benchmark built from the 50 questions in Razavi's two Analog Design Experiments With AI articles and graded by GPT-6 Luna against reference answers and rubrics, and use it to track how fast AI models improve at analog circuit analysis. Across 73 multimodal large language models released between May 2024 and November 2026, the top score rose from 28.83% to 98.33% in 30 months, mostly before the reference answers were published. The leading models now answer almost every textbook question correctly; the hardest questions show a circuit that resembles a textbook one but works differently, or ask about behavior the obvious analysis misses. Within two and a half years, a set of questions that exposed the limits of AI models has become one they answer almost perfectly.

查看此版本解读 →查看此版本原文 ↗