English(EN) SWE-Prime: Fewer Trajectories, Better Performance

新的SWE-Prime方法提高了LLM在软件工程任务上的性能

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-28 04:00

研究人员推出了一种新颖的两阶段方法SWE-Prime，用于选择数据来微调大型语言模型以完成软件工程任务。该方法在轨迹和片段级别过滤训练数据，以提高监督质量并减轻不良行为。在SWE-Bench Pro和SWE-Bench Verified上的实验表明，SWE-Prime精选的10%轨迹子集与使用完整的已解析数据集相比，可以带来显著的性能提升。 AI

影响该方法可能导致LLM在复杂的软件工程任务上进行更有效和更高效的训练，从而可能改进AI驱动的代码生成和调试。

排序理由该集群包含一篇详细介绍一种新方法以提高LLM性能的研究论文。[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: 26 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍一种新方法以提高LLM性能的研究论文。[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, model release; 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) · Dewu Zheng, Ruizhe Ye, Yanlin Wang, Yang Ye, Hongyu Zhang, Ensheng Shi, Xilin Liu, Yuchi Ma, Jianxing Yu, Zibin Zheng · 2026-08-28 04:00

SWE-Prime：轨迹更少，性能更优

arXiv:2608.27449v1 Announce Type: cross Abstract: To improve large language models' ability to resolve real-world software issues, prior work has focused on constructing large-scale agent trajectory datasets and performing supervised fine-tuning (SFT) on successful trajectories. …

本文如何被排名

报道来源 [1]

SWE-Prime：轨迹更少，性能更优

相关实体

相关话题