教大模型判断何时该停止思考，避免过度纠结。
CaRT: Teaching LLM Agents to Know When They Know Enough
- 用反事实轨迹对训练模型判断是否该终止
- 在医疗和数学任务中提升决策效率与成功率
- 适合需要审慎判断的交互式决策场景
许多任务要求模型通过多轮交互策略性地收集信息后再采取行动。这不仅需要有效获取信息的能力，还需判断何时停止收集并做出决策，以避免过度思考或行动偏差。本文提出反事实与推理终止（CaRT）方法，用于教会大模型何时停止信息搜集。CaRT通过反事实轨迹对进行微调：一条是应终止的轨迹，另一条是其最小修改版本且不应终止。模型需解释两种情况下的终止理由，通过语言推理能力实现判断机制的内化。我们在交互式医疗诊断和数学问题求解两个领域验证了该方法，在信息获取效率和任务成功率上均优于其他微调方法。
原文摘要 · Abstract (English)
Many tasks require learned models to strategically gather relevant information over multiple rounds of interaction before actually acting on a task. Strategic information gathering requires models to know not only how to effectively acquire information, but also when to stop gathering information and make a decision, in order to avoid overthinking or getting derailed when acting. In this paper, we formalize this problem and introduce Counterfactuals and Reasoning for Termination (CaRT), an approach for teaching LLMs when to stop seeking information. To appropriately learn when to terminate, CaRT fine-tunes LLMs using counterfactual pairs of trajectories, one where termination is appropriate and a minimally modified version of the same trajectory where it is not. It trains the LLM to explain the rationale for the termination decision in either case via verbal reasoning, and imbues this capability into the base LLM via fine-tuning. We instantiate CaRT in two domains: interactive medical diagnosis and math problem solving. In both domains, we find that CaRT improves the efficiency of information gathering and task success rate compared to other fine-tuning methods.
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。