PulseAugur
中
实时 00:00:31
English(EN) Mobile-Aptus: Confidence-Driven Proactive and Robust Interaction in MLLM-based Mobile-Using Agents

新框架提升MLLM移动代理的置信度和任务成功率

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员开发了Mobile-Aptus，一个置信度驱动的框架，旨在提高基于多模态大语言模型（MLLM）的移动代理的交互能力。该框架通过使代理能够输出置信度分数以及动作，然后使用语义相似性和直接偏好优化来纠正这些分数，从而解决了过度执行和过度请求的问题。Mobile-Aptus在四个基准测试中展示了最先进的性能，在任务成功率方面取得了显著的提高，并减少了对人工干预的需求。 AI

影响 增强了在移动设备上运行的AI代理的可靠性和效率，减少了不必要的人工干预。

排序理由 该集群包含一篇详细介绍MLLM代理新框架的研究论文。

在 arXiv cs.CL 阅读 →

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

新框架提升MLLM移动代理的置信度和任务成功率

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含一篇详细介绍MLLM代理新框架的研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
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
134 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.CL TIER_1 English(EN) · Zheng Wu, Pengzhou Cheng, Zongru Wu, Yuan Guo, Tianjie Ju, Aston Zhang, Gongshen Liu, Zhuosheng Zhang ·

    Mobile-Aptus：MLLM驱动的移动使用代理中的置信度驱动的主动和鲁棒交互

    arXiv:2605.28629v1 Announce Type: new Abstract: Recent advancements in multimodal large language models (MLLMs) have shown exceptional potential in enabling mobile-using agents to autonomously execute human instructions. However, fully automated agents often try to execute tasks …

  2. arXiv cs.CL TIER_1 English(EN) · Zhuosheng Zhang ·

    Mobile-Aptus：MLLM驱动的移动使用代理中的置信度驱动的主动鲁棒交互

    Recent advancements in multimodal large language models (MLLMs) have shown exceptional potential in enabling mobile-using agents to autonomously execute human instructions. However, fully automated agents often try to execute tasks even when they are unable to resolve them, leadi…

相关实体

相关话题