按年被提到
20232023 · 3 次2026
软件工程与 AI 编程›代码生成与程序合成·arXiv:2309.07062 · cs.PL

Large Language Models for Compiler Optimization

Chris Cummins, Volker Seeker, Dejan Grubisic, Mostafa Elhoushi, Youwei Liang, Baptiste Roziere, Jonas Gehring, Fabian Gloeckle, Kim Hazelwood, Gabriel Synnaeve, Hugh Leather · 2023
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 4 人首次 2023-09-14 · 最近 2024-06-18
4不同的人
2活跃月份
4讨论串
57被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We explore the novel application of Large Language Models to code optimization. We present a 7B-parameter transformer model trained from scratch to optimize LLVM assembly for code size. The model takes as input unoptimized assembly and outputs a list of compiler options to best optimize the program. Crucially, during training, we ask the model to predict the instruction counts before and after optimization, and the optimized code itself. These auxiliary learning tasks significantly improve the optimization performance of the model and improve the model's depth of understanding. We evaluate on a large suite of test programs. Our approach achieves a 3.0% improvement in reducing instruction counts over the compiler, outperforming two state-of-the-art baselines that require thousands of compilations. Furthermore, the model shows surprisingly strong code reasoning abilities, generating compilable code 91% of the time and perfectly emulating the output of the compiler 70% of the time.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论ec109685 · 2024-06

“There are similarities to the approach in this paper (though they trained a model from scratch): https://arxiv.org/pdf/2309.07062 How well would an LLM trained with a huge number of examples do on this test? Essentially with enough attention, Goodhart's law will take over.”

Getting 50% (SoTA) on Arc-AGI with GPT-4o原评论 ↗
被提交为 story · 3 次