根据模型能力动态选择评估方式，显著降低多智能体RAG的计算成本。
To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG
- 弱模型靠文档隔离解决上下文混淆，无需评分即可提效。
- 强模型需评分，提出无标签扰动探针识别其评分行为。
- 自适应路由架构MADARA零样本适配多模型，轻量高效。
多智能体文档评估在检索增强生成中计算开销大，促使实践者采用更小、可部署的模型，但其评估机制仍不清晰。我们在7B-9B指令微调模型上开展无训练干预的受控研究，覆盖多种QA基准，发现模型受益于评估的方式存在明显二分：对较弱基线，主要机制是单文档隔离；令人惊讶的是，无评分的隔离已能媲美完整多智能体评估，说明解决多文档上下文混淆比评分质量带来更大提升，最高达50个百分点。对于强基线，评分质量至关重要，我们引入基于无标签扰动的推理-评分耦合探针，以分类其评分行为。整合上述发现，提出模型自适应路由架构MADARA。关键在于，基于单一试点模型得出的诊断阈值可零样本泛化至四个未见模型族，提供鲁棒、轻量的流水线以消除计算开销。
原文摘要 · Abstract (English)
Multi-agent document assessment for retrieval-augmented generation is computationally expensive, driving practitioners toward smaller, deployable models whose assessment mechanisms remain poorly understood. We conduct a controlled study of training-free interventions on 7B-9B instruction-tuned models across diverse QA benchmarks, revealing a sharp dichotomy in how models benefit from assessment. For weaker baselines, the dominant mechanism is per-document isolation. Astoundingly, assessment-free isolation matches full multi-agent assessment, demonstrating that resolving multi-document context confusion, rather than scoring quality, drives outsized gains of up to 50 percentage points. Conversely, for strong baselines where scoring quality matters, we introduce Reasoning-Score Coupling, a label-free perturbation probe that classifies scoring behavior. Integrating these findings, we propose MADARA, a model-adaptive routing architecture. Crucially, MADARA's diagnostic thresholds derived from a single pilot model generalize zero-shot to four unseen model families, providing a robust, lightweight pipeline to eliminate computational overhead.
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。