English(EN) OpenMobile-2 proves open-source smartphone agents can rival proprietary models

开源智能手机代理通过多样化的应用训练可与专有模型相媲美

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-25 06:55

研究人员开发了 OpenMobile-2，这是一个用于智能手机代理的开源训练环境，它利用了 110 多个应用程序，包括流行应用程序的副本。这种方法侧重于训练多样性而非数据量，已产生了一个开源代理，其创建者声称该代理在真实的 Android 设备上的表现优于其他开源代理。 AI

影响这一发展表明，开源代理可以通过多样化的应用程序训练来实现与专有模型相媲美的性能。

排序理由该项目描述了一个用于智能手机代理的新开源训练环境和模型。[lever_c_demoted from research: ic=1 ai=1.0]

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: 32 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该项目描述了一个用于智能手机代理的新开源训练环境和模型。[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: model release, 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。

The Decoder TIER_1 English(EN) · Jonathan Kemper · 2026-10-25 06:55

OpenMobile-2 证明开源智能手机代理可与专有模型相媲美

<p><img alt="A smartphone surrounded by icons (camera, shopping, music, video, calendar, cloud, map) on a blue background." class="attachment-full size-full wp-post-image" height="1047" src="https://the-decoder.com/wp-content/uploads/2026/10/open-mobile-2-smartphone-app-training-…