核心要点
- 问题/背景
- 这篇论文把 agent 从被动响应用户显式请求，推进到主动发现上下文中尚未被用户指出的多个隐藏问题。
- 方法/机制
- TIDE 的任务定义要求 agent 从文档、工具或代码上下文中发现多个并存问题，给出支持证据，并配套具体行动建议。
- 结果/证据
- 方法上，它用 iterative discovery 避免单轮输出只锚定最显著问题，再用 thought templates 作为可复用 schema，引导模型关注特定上下文信号并连接到问题类型。
- 收录价值
- 收录价值在于它把 proactive agents 的一个关键能力问题化：不是等用户提问，而是主动覆盖上下文中的未知问题集合。这对 coding agents、personal workspace agents 和治理/审计 agent 都有通用价值。
原始摘要与中文对照
中文对照翻译
模型假设用户已经知道哪里出了问题以及该问什么。然而，在实践中，最具影响的问题往往是用户尚未注意到的：口头批准但尚未书面记录的预算，导致供应商订单在严格的截止日期前受阻；同一报告的两份副本数字冲突，都将提交给即将进行的审查；团队心照不宣地停止参加的定期会议，仍然占据着紧急启动的唯一窗口。此类问题通常显而易见地存在于代理原则上可以检查的文档、电子邮件和日历条目中。这些看似不同的问题具有共同的结构，这种结构超越了上述工作空间设置。在代理运行的数字环境中（个人工作空间、软件仓库或其他丰富的工作上下文），证据不断积累，其中多个问题并存；没有一个问题被明确表述为请求；并存问题的数量事先未知；并且只解决最突出的问题会使其余问题未被触及。因此，我们认为主动协助最好不要被框定为预测单一用户意图，而是作为从上下文中发现多个隐藏问题的更广泛任务。现有关于主动代理的工作研究了何时进行干预（Liu et al., 2025; Zhang et al., 2025）或如何预测单一局部需求（Lu et al., 2025; Yang et al., 2025a; Pasternak et al., 2025），但在很大程度上回避了真实工作流所需的多问题、上下文范围的设置。完成这项任务需要两种互补的能力：对与更突出问题争夺注意力的并存问题进行广泛覆盖，以及每个候选问题具有足够的精确度以使其可操作而非推测性。为了应对这些挑战，我们提出了TIDE（Template-guided Iterative Discovery and rEsolution；图1），这是一个结合了沿不同轴线运行的两种机制的框架。具体而言，代理作为文档、工具和代码的助手被广泛部署。然而，它们通常只根据明确的用户请求采取行动，这只会揭示用户已经注意到的问题，而许多其他重要问题则并存，显而易见地隐藏在更广泛的用户上下文中，其总数事先未知。我们将此框定为从上下文中发现多个隐藏问题的任务，其中并存问题应被揭示，以支持证据为基础，并与具体行动相结合。为此，我们引入了TIDE，一个具有两种互补机制的模板引导迭代框架。具体而言，受单次预测锚定于最突出案例并产生通用主张的观察启发，我们提出了迭代发现，它在每一轮中揭示一小批候选问题，同时以已发现的内容为条件，从而使后续轮次扩展覆盖范围；以及思维模板，这是从先前解决的案例中提炼出的可重用模式，它们指定要关注哪些上下文信号以及如何连接它们，将每个预测锚定在一个可识别的问题类别中。我们在两个现实设置（个人工作空间和软件仓库）上，使用四种骨干模型验证了TIDE，结果显示在任务覆盖、识别和解决方面，TIDE相对于单次和并行多代理基线有显著提升。
原始摘要
model presumes that the user already knows what is wrong and what to ask. In practice, however, the most consequential issues are often the ones a user has not yet noticed: a budget approval given verbally but not yet recorded in writing, holding up a vendor order on a hard deadline; two copies of the same report with conflicting numbers, both heading into an upcoming review; a recurring meeting the team has tacitly stopped attending, still blocking the only window for an urgent kickoff. Such issues sit, often in plain sight, inside the very documents, emails, and calendar entries that the agent could in principle inspect. These otherwise different issues share a common structure that extends beyond the workspace setting above. Across the digital environments where agents operate (a personal workspace, a software repository, or another rich working context), evidence accumulates in which multiple problems coexist; none is articulated as a request; the number of coexisting problems is not known in advance; and resolving only the most salient one leaves the rest untouched. We therefore argue that proactive assistance is best framed not as anticipating a single user intent, but as the broader task of discovering multiple hidden problems from context. Existing work on proactive agents has studied when to intervene (Liu et al., 2025; Zhang et al., 2025) or how to anticipate a single localized need (Lu et al., 2025; Yang et al., 2025a; Pasternak et al., 2025), but largely sidesteps this multi-problem, contextwide setting that real workflows demand. Meeting this task requires two complementary capabilities: broad coverage over coexisting problems that compete for attention with more salient ones, and enough precision per candidate to be actionable rather than speculative. To address these challenges, we present TIDE (Template-guided Iterative Discovery and rEsolution; Figure 1), a framework that combines two mechanisms operating along distinct axes. Specifically, motivated Agents are widely deployed as assistants over documents, tools, and code. However, they typically act only on explicit user requests, which surface only the problems the user has noticed, while many other important problems coexist, hidden in plain sight, within the broader user context, with their total number unknown in advance. We frame this as the task of discovering multiple hidden problems from context, in which coexisting problems should be uncovered, grounded in supporting evidence, and paired with concrete actions. To this end, we introduce TIDE, a template-guided iterative framework with two complementary mechanisms. Specifically, motivated by the observation that single-pass prediction anchors on the most salient cases and yields generic claims, we propose iterative discovery, which surfaces a small batch of candidates per round while conditioning on what has already been found, so subsequent rounds extend coverage; and thought templates, reusable schemas distilled from previously solved cases that specify what contextual signals to attend to and how to connect them, anchoring each prediction in a recognizable problem class. We validate TIDE on two realistic settings, personal workspaces and software repositories, across four model backbones, showing substantial gains over singleshot and parallel multi-agent baselines on task coverage, identification, and resolution.