跳到正文
arXiv cs.AI· Xiaoyang Xing, Wenyu Dai, Xiaoliang Gong, Asoke K Nandi, Anthony Cohn·· 10 小时前AI 评分31

L2C:LLM 引导的多智能体工作流演化,实现低 token 红队测试

Learning to Coordinate: LLM-Guided Workflow Evolution for Token-Efficient Red Teaming

AI 导读

研究者提出 Learning to Coordinate(L2C),一个由 LLM 引导、为每个红队测试案例演化可执行多智能体工作流的框架,用版本化图表示智能体职责与依赖,并由经监督微调的控制器决定保留、改写或重构协作。

正文 · 原文

View PDF HTML (experimental)

Abstract:Automated red teaming is essential for identifying safety failures in large language models (LLMs), but its practical scalability depends on the resources required to discover those failures. Multi-agent systems can combine complementary capabilities, yet repeated planning, redundant exchanges, and ineffective information routing can make their coordination expensive. Moreover, a target model's response indicates the outcome of a test without identifying which collaboration decision should change. We present Learning to Coordinate (L2C), an LLM-guided framework that evolves an executable multi-agent workflow for each red-teaming case. L2C represents agent responsibilities, execution dependencies, and input bindings in a versioned graph, and associates each graph with records of the information actually delivered and produced. A controller trained through supervised fine-tuning integrates these records with target responses, evaluator verdicts, and remaining budget to retain the workflow, refine instructions, restructure collaboration, or lock further editing. This design connects feedback to explicit changes in subsequent execution while allowing a useful organization to continue without repeated controller intervention. The evaluation protocol covers HarmBench with three execution backbones and two target models, together with component ablations, dataset transfer, and evaluator checks. Across the six executor-target configurations, L2C improves ASR over EvoSynth by 1.4 to 4.5 percentage points while reducing mean online tokens by 35.6% to 37.3%. L2C connects adaptive collaboration to the resource requirements of automated security evaluation.
Comments: 10 pages, 3 figures
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.24221 [cs.AI]
  (or arXiv:2611.24221v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.24221

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Xiaoyang Xing [view email]
[v1] Wed, 18 Nov 2026 10:58:44 UTC (3,309 KB)

来源:arXiv cs.AI · arxiv.org