Princeton 与 Google 论文：4B 小模型接管智能体决策，成本降 81%-89%
Princeton 和 Google 的论文，用 4B 小模型替大模型做智能体的工具选择，成本砍掉八成多，速度还快 2.5 倍，做 Agent 的都该看看。
Princeton 大学与 Google 的论文 JevAgentHarness 指出，在孤立问题上最准的决策模型在智能体任务中只解决 69.5% 的任务，而在智能体实际状态上训练的 AgentDecision-4B 达到 91.5%。该框架让小模型回答每一步的多选题（如选哪个工具），只在置信度超过阈值时采纳其答案。在匹配训练数据的任务上，这种方式让智能体便宜 81.1% 到 89.0%，速度快至 2.5 倍，成功率波动在 3 个点以内。相比之下，不加阈值地把所有工具选择都交给小模型，成功率会下降 19.0 个点。
New Princeton Univ + Google paper finds that the decision model most accurate on isolated questions solved only 69.5% of agent tasks, while a 4B model trained on agent states solved 91.5%.
Agent harnesses call a big LLM at every step, which makes long tasks slow and costly. Yet most of those calls just pick from a list, like which tool to use next.
Their harness asks a fast decision model each multiple-choice question and uses its answer only above a confidence threshold fitted on the agent's own runs. They also trained AgentDecision-4B on states the agent actually reaches, including after its own mistakes.
On tasks like their training data, 4B models trained this way made agents 81.1% to 89.0% cheaper and up to 2.5x faster. Success stayed within 3 points, while handing over every tool choice with no threshold lowered success by 19.0 points.
To hand routine agent choices to a small model, train it and set its confidence threshold on your agent's own runs.
– arxiv. org/abs/2610.21412
Title: "JevAgentHarness: Benchmarking, Deploying and Training Decision Models That Share Control with LLMs"
