基于源文档的知识图谱-RAG：用于长文档问答的 grounded 知识图谱
GROUNDEDKG-RAG: Grounded Knowledge Graph Index for Long-document Question Answering
AI总结：
本文提出 groundedkg-rag，通过从源文档中显式提取并 grounding 知识图谱，提升长文档问答的效率和事实准确性，实验表明其在成本较低的情况下性能优于现有基线。
摘要：
检索增强生成（RAG）系统因其能提升生成质量并减少所需输入上下文长度而被广泛应用于现代大型语言模型（LLMs）。本文聚焦于长文档问答中的 RAG 系统。当前方法依赖于 LLM 描述，导致资源消耗高、延迟大、层次间内容重复以及由于源文本缺乏或有限 grounding 而产生的幻觉。为提高效率和事实准确性，我们提出了 GroundedKG-RAG，一种知识图谱显式提取并 grounding 在源文档中的 RAG 系统。具体而言，我们将 GroundedKG 的节点定义为实体和动作，边定义为时间或语义关系，每个节点和边都 grounding 在原始句子中。我们通过语义角色标注（SRL）和抽象意义表示（AMR）解析构建 GroundedKG 并对其进行嵌入。在查询过程中，我们对查询应用相同的转换，从 grounded 源文本中检索最相关的句子以进行问答。我们在 NarrativeQA 数据集的示例上评估了 GroundedKG-RAG，发现其在成本较低的情况下性能与最先进的专有长上下文模型相当，并优于竞争基线。此外，我们的 GroundedKG 可被人类解释和阅读，有助于结果审核和错误分析。
英文摘要：
Retrieval-augmented generation (RAG) systems have been widely adopted in contemporary large language models (LLMs) due to their ability to improve generation quality while reducing the required input context length. In this work, we focus on RAG systems for long-document question answering. Current approaches suffer from a heavy reliance on LLM descriptions resulting in high resource consumption and latency, repetitive content across hierarchical levels, and hallucinations due to no or limited grounding in the source text. To improve both efficiency and factual accuracy through grounding, we propose GroundedKG-RAG, a RAG system in which the knowledge graph is explicitly extracted from and grounded in the source document. Specifically, we define nodes in GroundedKG as entities and actions, and edges as temporal or semantic relations, with each node and edge grounded in the original sentences. We construct GroundedKG from semantic role labeling (SRL) and abstract meaning representation (AMR) parses and then embed it for retrieval. During querying, we apply the same transformation to the query and retrieve the most relevant sentences from the grounded source text for question answering. We evaluate GroundedKG-RAG on examples from the NarrativeQA dataset and find that it performs on par with a state-of-the art proprietary long-context model at smaller cost and outperforms a competitive baseline. Additionally, our GroundedKG is interpretable and readable by humans, facilitating auditing of results and error analysis.
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。
