arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.01626cs.CLcs.AIcs.DBcs.IR
收藏

SPARKLE:在解码中直接集成知识图谱以增强SPARQL生成

SPARKLE: Enhancing SPARQL Generation with Direct KG Integration in Decoding

Samsung Research(三星研究院)

Jaebok Lee, Hyeonjeong Shin

AI总结:

SPARKLE提出一种端到端NL2SPARQL框架,在解码中直接集成知识图谱结构,显著减少无效查询生成,并在多个KBQA基准上取得最优或接近最优结果。

摘要:

现有的KBQA方法通常依赖多阶段流程,涉及实体链接、子图检索和查询结构生成等任务。然而,多阶段方法依赖于前序步骤的准确性,容易导致级联错误并增加推理时间。尽管少数研究探索了端到端模型,但它们往往准确率较低,并会生成底层数据不支持的无效查询。此外,大多数先前方法局限于静态训练数据,可能忽视了知识库随时间的演化特性。为应对这些挑战,我们提出了一种新颖的端到端自然语言到SPARQL框架——SPARKLE。值得注意的是,SPARKLE在解码过程中直接利用知识库的结构,将知识有效集成到查询生成中。我们的研究表明,仅在推理阶段引用知识库就能显著减少不可执行查询的生成。SPARKLE在SimpleQuestions-Wiki上取得了新的最优结果,并在LCQuAD 1.0上获得了最高F1分数(在不使用金标实体的模型中),而在WebQSP数据集上的结果略低。最后,我们展示了SPARKLE的快速推理速度,以及其在训练与推理阶段知识库不同时的适应能力。

英文摘要:

Existing KBQA methods have traditionally relied on multi-stage methodologies, involving tasks such as entity linking, subgraph retrieval and query structure generation. However, multi-stage approaches are dependent on the accuracy of preceding steps, leading to cascading errors and increased inference time. Although a few studies have explored the use of end-to-end models, they often suffer from lower accuracy and generate inoperative query that is not supported by the underlying data. Furthermore, most prior approaches are limited to the static training data, potentially overlooking the evolving nature of knowledge bases over time. To address these challenges, we present a novel end-to-end natural language to SPARQL framework, SPARKLE. Notably SPARKLE leverages the structure of knowledge base directly during the decoding, effectively integrating knowledge into the query generation. Our study reveals that simply referencing knowledge base during inference significantly reduces the occurrence of inexecutable query generations. SPARKLE achieves new state-of-the-art results on SimpleQuestions-Wiki and highest F1 score on LCQuAD 1.0 (among models not using gold entities), while getting slightly lower result on the WebQSP dataset. Finally, we demonstrate SPARKLE's fast inference speed and its ability to adapt when the knowledge base differs between the training and inference stages.

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。

↑