用多臂老虎机动态选择检索策略，让知识图谱增强的AI更适应变化环境。
Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs
- 将不同检索方法视为可选'臂'，用实时反馈动态调整
- 在非平稳环境下性能超越基线，平稳时也达顶尖水平
- 适合需要持续更新、对响应速度敏感的智能问答场景
尽管大语言模型在诸多自然语言任务中表现优异，但在记忆海量世界知识方面仍存在显著局限。近期研究表明，结合知识图谱（Knowledge Graphs）与检索增强生成（RAG）框架，能有效提升大模型的推理能力。然而，在真实场景部署时面临挑战：非平稳环境持续演化可能导致性能下降，同时需平衡性能与响应速度以保障用户体验。为此，本文提出一种多目标多臂老虎机增强的RAG框架，支持多种具备不同能力的检索方法，在复杂多变的检索环境中协同工作。每个检索方法被视为一个独立的“臂”，系统根据输入查询及各臂历史多目标表现，结合实时用户反馈，动态选择最优检索策略。在两个基准知识图谱问答（KGQA）数据集上的大量实验表明，该方法在非平稳设置下显著优于基线，在平稳环境下亦达到当前最优性能。代码与数据已公开于 https://github.com/FUTUREEEEEE/Dynamic-RAG.git。
原文摘要 · Abstract (English)
Despite the superior performance of Large language models on many NLP tasks, they still face significant limitations in memorizing extensive world knowledge. Recent studies have demonstrated that leveraging the Retrieval-Augmented Generation (RAG) framework, combined with Knowledge Graphs that encapsulate extensive factual data in a structured format, robustly enhances the reasoning capabilities of LLMs. However, deploying such systems in real-world scenarios presents challenges: the continuous evolution of non-stationary environments may lead to performance degradation and user satisfaction requires a careful balance of performance and responsiveness. To address these challenges, we introduce a Multi-objective Multi-Armed Bandit enhanced RAG framework, supported by multiple retrieval methods with diverse capabilities under rich and evolving retrieval contexts in practice. Within this framework, each retrieval method is treated as a distinct ``arm''. The system utilizes real-time user feedback to adapt to dynamic environments, by selecting the appropriate retrieval method based on input queries and the historical multi-objective performance of each arm. Extensive experiments conducted on two benchmark KGQA datasets demonstrate that our method significantly outperforms baseline methods in non-stationary settings while achieving state-of-the-art performance in stationary environments. Code and data are available at https://github.com/FUTUREEEEEE/Dynamic-RAG.git
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。