让大模型智能判断何时该停止行动，避免无效操作。
Agentic Abstention: Do Agents Know When to Stop Instead of Act?
- 设计了多轮交互中的停顿决策机制，解决模型盲目执行问题。
- 在28000+任务上测试，发现多数模型或过早、或过晚选择放弃。
- 提出CONVOLVE方法，无需改模型即可显著提升停顿准确率。
大型语言模型代理需通过多轮交互完成用户目标，但并非所有目标都明确或可实现。可靠的代理应在不确定时识别出继续操作无益，并主动停止。我们提出「代理式回避」（Agentic Abstention），即在不确定性下决定是否停止的序列决策问题。不同于单轮回答或放弃的判定，代理需在每步选择回答、放弃或获取更多信息，且放弃时机可能仅在环境反馈后才显现。我们在网页购物、终端环境和问答任务中评估了13个基于LLM的代理系统及2个代理框架，覆盖超过28,000个任务。结果表明，关键挑战不仅在于能否放弃，更在于何时放弃：部分代理从不放弃，另一些则在多次无效操作后才放弃。该差距在指令看似可行但环境揭示不可行的任务中尤为明显（如无匹配结果）。我们还发现模型规模、推理能力与代理架构对放弃行为影响各异，更大的模型有时反而更迟钝。最后，我们提出CONVOLVE，一种上下文工程方法，将完整交互轨迹提炼为可复用的停止规则。在WebShop上，它无需更新参数即可将Llama-3.3-70B的及时召回率从26.7%提升至57.4%。数据集与代码已公开。
原文摘要 · Abstract (English)
LLM agents are expected to act over multiple turns, using search, browsing interfaces, and terminal tools to complete user goals. Yet not every goal is well specified or achievable in the available environment. In such cases, a reliable agent should recognize that further interaction is unlikely to help and abstain from additional tool calls. We define Agentic Abstention, the problem of deciding when an agent should stop acting under uncertainty. Unlike standard LLM abstention, which is usually evaluated as a single-turn answer-or-abstain decision, agentic abstention is a sequential decision problem: an agent can answer, abstain, or gather more information at each turn, and the need to abstain may only become clear after interacting with the environment. We study this problem across web shopping, terminal environments, and question answering, evaluating 13 LLM-as-agent systems and 2 agent scaffolds on more than 28,000 tasks. Our results show that the main challenge is not only whether agents can abstain, but also when they abstain. Some agents never abstain when they should, while others do so only after many unnecessary interactions. This gap is especially large on tasks where the instruction appears feasible until the environment reveals otherwise (e.g., no valid result matches the instruction). We further find that model scale, reasoning, and agent scaffolding affect abstention in different ways, where larger or more capable models sometimes perform worse at timely abstention. Finally, we introduce CONVOLVE, a context engineering method for improving agentic abstention that distills full interaction trajectories into reusable stopping rules. On WebShop, CONVOLVE substantially improves timely abstention without updating model parameters, raising Llama-3.3-70B's timely recall rate from 26.7 to 57.4. Our dataset and code are available at https://lhannnn.github.io/agentic-abstention
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。