English(EN) AdaExplore: Failure-Driven Adaptation and Diversity-Preserving Search for Efficient Kernel Generation

新AI代理AdaExplore提升内核代码生成效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-10 04:00

研究人员开发了AdaExplore，一个旨在提高高效内核代码生成能力的创新代理框架，特别适用于Triton等领域特定语言。该系统采用两阶段流程：面向失败的自适应，它从反复出现的错误中构建有效性规则的记忆；以及保持多样性的搜索，它通过在局部改进和结构化再生之间交替来探索优化空间。在KernelBench基准测试上的实验表明，AdaExplore在100步内分别在Level-2和Level-3基准测试上实现了3.11倍和1.62倍的显著加速。 AI

影响 AdaExplore的面向失败的自适应和保持多样性的搜索方法可以加速用于代码生成任务的专业AI代理的开发。

排序理由发布了一篇详细介绍新AI框架及其基准测试结果的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 13 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 发布了一篇详细介绍新AI框架及其基准测试结果的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, product; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Weihua Du, Jingming Zhuo, Yixin Dong, Andre Wang He, Weiwei Sun, Zeyu Zheng, Manupa Karunaratne, Ivan Fox, Tim Dettmers, Tianqi Chen, Yiming Yang, Sean Welleck · 2026-09-10 04:00

AdaExplore：面向高效内核生成的失败驱动自适应与多样性保持搜索

arXiv:2604.16625v2 Announce Type: replace-cross Abstract: Recent large language model (LLM) agents have shown promise in using execution feedback for test-time adaptation. However, robust self-improvement remains far from solved: most approaches still treat each problem instance …

本文如何被排名

报道来源 [1]

AdaExplore：面向高效内核生成的失败驱动自适应与多样性保持搜索

相关实体

相关话题