ProactiveMobile：面向移动设备主动式智能增强的综合基准
ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence On Mobile Devices
为视觉语言模型提供了一个可执行、可客观评测的主动式移动端任务基准：模型需从屏幕截图等设备上下文信号中推断潜在用户意图，并输出由 63 个 API 组成的可执行函数序列，可直接用于评测 VLM 的意图推理与工具调用能力。
原文依据 ProactiveMobile formalizes the proactive task as inferring latent user intent across four dimensions of on-device contextual signals
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多模态大语言模型（MLLM）在移动智能体开发中已取得显著进展，但其能力主要局限于被动式范式，即仅执行用户明确下达的指令。新兴的主动式智能范式——智能体自主预判需求并主动发起动作——代表了移动智能体的下一个前沿方向。然而，该方向的发展受到关键瓶颈的制约：缺乏能够应对真实世界复杂性、并支持客观且可执行评测的基准。为克服这些挑战，我们提出 ProactiveMobile，一个旨在系统性推进该领域研究的综合基准。ProactiveMobile 将主动式任务形式化为：在四个维度的设备端上下文信号中推断潜变量用户意图，并从包含 63 个 API 的综合函数池中生成可执行的函数序列。该基准包含 14 个场景的 3,660 余条实例，并通过多答案标注来体现真实世界的复杂性。为确保质量，由 30 名专家组成的团队对基准进行最终审核，核验事实准确性、逻辑一致性与动作可行性，并修正不合规条目。大量实验表明，我们微调后的 Qwen2.5-VL-7B-Instruct 成功率达到 20.82%，优于 o1（17.02%）与 GPT-5（11.37%）。该结果表明，主动式能力是当前 MLLM 普遍缺失的一项关键能力，但它是可学习的，这凸显了所提出基准在主动式能力评测方面的重要性。数据与代码见 https://github.com/xiaomi-research/proactive-mobile。
英文原文摘要 1683 字符
Multimodal large language models (MLLMs) have made significant progress in mobile agent development, yet their capabilities are predominantly confined to a reactive paradigm, where they merely execute explicit user commands. The emerging paradigm of proactive intelligence, where agents autonomously anticipate needs and initiate actions, represents the next frontier for mobile agents. However, its development is critically bottlenecked by the lack of benchmarks that can address real-world complexity and enable objective, executable evaluation. To overcome these challenges, we introduce ProactiveMobile, a comprehensive benchmark designed to systematically advance research in this domain. ProactiveMobile formalizes the proactive task as inferring latent user intent across four dimensions of on-device contextual signals and generating an executable function sequence from a comprehensive function pool of 63 APIs. The benchmark features over 3,660 instances of 14 scenarios that embrace real-world complexity through multi-answer annotations. To ensure quality, a team of 30 experts conducts a final audit of the benchmark, verifying factual accuracy, logical consistency, and action feasibility, and correcting non-compliant entries. Extensive experiments demonstrate that our fine-tuned Qwen2.5-VL-7B-Instruct achieves a success rate of 20.82%, outperforming o1 (17.02%) and GPT-5 (11.37%). This result indicates that proactivity is a critical competency widely lacking in current MLLMs, yet it is learnable, emphasizing the importance of the proposed benchmark for proactivity evaluation. Data and code are available at https://github.com/xiaomi-research/proactive-mobile.