arXiv：cs.LG· Ritik Raj, Souvik Kundu, Dheemanth Joshi, Tushar Krishna·· 5 小时前AI 评分41
ORACLE：通过自适应验证器校准反馈的智能体 AI 编排路由
ORACLE: Agentic AI Orchestrator Routing Via Adaptive Verifier Calibration Feedback
AI 导读
ORACLE 是一种并发感知的在线路由机制，将自适应路由与自适应验证对齐，作为无需训练的反馈回路叠加在任意模型选择策略上，先分类任务类型再动态分配适配的验证器。在 SWE-bench、tau2-bench 和 Terminal-Bench 2.0 上，ORACLE 将准确率-成本前沿较 SOTA 路由基线最多提升 7 个百分点，配合 DISC 调度器可将程序吞吐量提升至多 1.8 倍。代码已开源。
正文
Abstract:Modern enterprise agent deployments consist of a heterogeneous pool of large language models (LLMs) having diverse capabilities and cost. Existing model routing strategies optimize the quality-cost trade-off, while providing request-level static decisions. More recent solutions address agentic routing as a task-level selection with a serial verifier based router feedback loop. However, their fixed verifier suitable for homogeneous workloads may not generalize to heterogeneous batches of agentic tasks (example: coding, general conversational). Additionally, due to the verifier placement in the critical path of the loop, serving quality may be affected during multiple concurrent requests routing. To mitigate these issues, we present ORACLE. It is a concurrency-aware online routing mechanism that aligns adaptive routing with adaptive verification for feedback. ORACLE acts as a training-free drop-in 'feedback loop' on top of any model-selection policy to first classify the task type and then dynamically assigns a task-appropriate verifier. Further, we develop a delayed feedback strategy for concurrent requests that largely removes verifier latency from the dispatch critical path. We then present a post-routing dispatch scheduler, namely DISC. DISC reserves each task's peak KV footprint at admission and dispatches to an alternate backend when the reward gain from reduced wait exceeds the reward loss from lower accuracy. Extensive evaluation on SWE-bench, tau2-bench, and Terminal-Bench 2.0 shows that ORACLE improves the accuracy-cost frontier by up to 7 percentage points over state-of-the-art routing baselines, while ORACLE with DISC improves program throughput by up to 1.8x. To facilitate reproducibility and support future development, we open-source the code-base of ORACLE at this https URL.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2607.22465 [cs.AI] |
| (or arXiv:2607.22465v4 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.22465 arXiv-issued DOI via DataCite |
Submission history
From: Ritik Raj [view email]
[v1]
Fri, 24 Jul 2026 16:29:06 UTC (1,196 KB)
[v2]
Mon, 27 Jul 2026 17:28:05 UTC (1,196 KB)
[v3]
Tue, 29 Sep 2026 02:30:26 UTC (265 KB)
[v4]
Thu, 1 Oct 2026 21:34:02 UTC (265 KB)
来源：arXiv：cs.LG · arxiv.org