PulseAugur
实时 19:44:20

新的GRTO框架将强化学习与可微分工具使用相结合,用于分割

研究人员开发了一个名为Group Relative Tool Optimization (GRTO)的新框架,以改进计算机视觉中的指代分割任务。该方法将强化学习与可微分工具使用相结合,允许分割解码器与主策略一起进行优化。一种预训练技术Bootstrapped-GRTO (B-GRTO)进一步提高了收敛速度和性能。实验表明,B-GRTO在具有挑战性的分割基准测试中显著优于现有方法。 AI

影响 引入了一种将强化学习与可微分工具使用相结合的新颖方法,有望提高复杂视觉-语言分割任务的性能。

排序理由 该集群包含一篇详细介绍新研究方法的学术论文。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. arXiv cs.LG TIER_1 · Mario Markov (INSAIT, Sofia University "St. Kliment Ohridski"), Stefan Maria Ailuro (INSAIT, Sofia University "St. Kliment Ohridski"), Mohammad Mahdi (INSAIT, Sofia University "St. Kliment Ohridski"), Luc Van Gool (INSAIT, Sofia University "St. Kliment O… ·

    B-GRTO: Bootstrapped Group Relative Tool Optimization for Referring Segmentation

    arXiv:2605.23500v1 Announce Type: cross Abstract: Segmentation is a fundamental task in computer vision, underpinning pixel-level scene understanding and serving as a cornerstone for applications ranging from autonomous perception to medical image analysis. For complex referring …

  2. arXiv cs.CV TIER_1 · Danda Pani Paudel ·

    B-GRTO: Bootstrapped Group Relative Tool Optimization for Referring Segmentation

    Segmentation is a fundamental task in computer vision, underpinning pixel-level scene understanding and serving as a cornerstone for applications ranging from autonomous perception to medical image analysis. For complex referring segmentation, recent methods pair large vision-lan…