arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.25985cs.CRcs.AIcs.LG
收藏

NIOCIS:面向工具使用智能体的带节点条件选项空间的类型化决策下一跳路由

NIOCIS: Typed-Decision Next-Hop Routing with Node-Conditioned Option Spaces for Tool-Use Agents

Jin Han, Mingfeng Xin, Feng Liang, Jingtao Yu

AI总结:

本研究提出NIOCIS路由层,将执行日志转化为循环折叠动作图,在安全运营场景中验证其可替换LLM下一跳决策,性能受状态特异性等因素限制,为工具使用智能体提供低成本路由方案。

摘要:

工具使用智能体的每一次决策都需付费:每一跳时,语言模型会重新读取状态并思考数秒,尽管其动作有限且决策会重复出现。我们提出NIOCIS,这是一个将执行日志折叠为循环折叠动作图(包含前向转移及恢复回边)的路由层,用于回答一个类型化问题:在某个状态下,工具集群(fleet)接下来会选择哪个已声明选项,加上停止动作。一个小型模型在一次传递中读取这些选项的校准分布,不生成任何文本。在1615个生产安全运营场景中,我们测量该层所替换的部分、返回的部分及其边界。替换部分:在3875个保留决策中,类型化编码器与日志记录的下一跳匹配率为32.2%(在已声明选项内为34.9%),编码器耗时10毫秒(类型化选择形式为19.5毫秒),而每个LLM跳的耗时为2.2秒;两个特征读数分别达到32.8%和32.3%,配对测试未区分出优劣,因此选择类型化层是出于成本考量,而非性能领先。调优部分:jpt-4b的零样本准确率为9.9%,微调后为31.9%,精确配对测试未区分出该差距(p=0.68),耗时244毫秒。返回部分:7.9%的决策中实际跳落在已声明选项之外,9.3%的决策中实际跳落在模型前20之外;最具置信度的15%决策准确率达91.9%。边界部分:跨轨迹留一法测试显示,随着状态键变得更具体,已发布状态的上限为25.6%至44.9%,即重复状态的分歧频率高于一致频率;选项空间消融实验表明,性能提升源于每节点候选限制(比平面目录高3.5个百分点),而非图遍历(区间内无提升);门控自演化循环使每个分支均高于其基准,但仍低于全监督;置信度门控级联在升级分支作为零样本替代时会失去一致性;且工具集群超过三分之一的一致性需要状态携带计划和观测。

英文摘要:

Tool-use agents pay per decision: at each hop a language model re-reads the state and thinks for seconds, though its actions are finite and decisions recur. We present NIOCIS, a routing layer that folds execution logs into a loop-collapsed action graph (forward transitions plus recovery back-edges) answering one typed question: which declared option at a state, plus stop, does the fleet take next? A small model reads a calibrated distribution over them in one pass, emitting no text. On 1,615 production security-operations episodes we measure what the layer replaces, returns and is bounded by. Replaces: the typed encoder matches the logged next hop on 32.2% of the 3,875 held-out decisions (34.9% within the declared options) in 10 ms of encoder time (19.5 ms for the typed-choice form), against 2.2 s per LLM hop; two feature readouts reach 32.8% and 32.3%, an ordering the paired test does not resolve, so the typed layer is chosen for cost, not a separated lead. Tuning: jpt-4b reaches 9.9% zero-shot and 31.9% fine-tuned, a gap an exact paired test does not resolve (p=0.68), at 244 ms. Returns little: the realised hop falls outside the declared options for 7.9% of decisions and outside the model's top twenty for 9.3%; the most confident 15% holds 91.9% accuracy. Bounded: a cross-trace leave-one-out test puts the released state's ceiling at 25.6--44.9% as the state key grows more specific, i.e. repeated states disagree more often than they agree; the option-space ablation attributes the gain to per-node candidate restriction (+3.5 points over a flat catalog) rather than graph traversal (zero within its interval); a gated self-evolution loop lifts every arm above its anchor yet ends below full supervision; a confidence-gated cascade loses agreement when the escalation arm is the zero-shot stand-in; and agreement past a third of the fleet needs a state carrying the plan and the observation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑