新研究通过路径监督和校准改进知识图谱问答

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-08 17:57

两篇新研究论文介绍了改进知识图谱问答（KGQA）的新方法。第一篇，PathISE，专注于从答案级别的标签中学习信息性路径监督，以训练模型从知识图谱中检索相关证据。第二篇，Conformal Path Reasoning（CPR），通过使用共形预测进行路径级别的校准来增强可信度，确保覆盖率的同时减小预测集的大小。 AI

影响这些方法旨在使知识图谱问答更加准确和可靠，从而可能改善用户与结构化数据的交互方式。

排序理由 arXiv上发表的两篇学术论文提出了新的KGQA方法。

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.AI TIER_1 English(EN) · Jianzhong Qi · 2026-05-11 16:21

PathISE: Learning Informative Path Supervision for Knowledge Graph Question Answering

Knowledge Graph Question Answering (KGQA) aims to answer user questions by reasoning over Knowledge Graphs (KGs). Recent KGQA methods mainly follow the retrieval-augmented generation paradigm to ground Large Language Models~(LLMs) with structured knowledge from KGs. However, trai…
arXiv cs.CL TIER_1 English(EN) · Dimitris N. Metaxas · 2026-05-08 17:57

Conformal Path Reasoning: Trustworthy Knowledge Graph Question Answering via Path-Level Calibration

Knowledge Graph Question Answering (KGQA) has shown promise for grounded and interpretable reasoning, yet existing approaches often fail to provide reliable coverage guarantees over retrieved answers. While Conformal Prediction (CP) offers a principled framework for producing pre…

报道来源 [2]

PathISE: Learning Informative Path Supervision for Knowledge Graph Question Answering

Conformal Path Reasoning: Trustworthy Knowledge Graph Question Answering via Path-Level Calibration

相关实体

相关话题