Abstract:Knowledge graph reasoning (KGR) aims to infer new knowledge or detect noises, which is essential for improving the quality of knowledge graphs. Recently, various KGR techniques, such as symbolic- and embedding-based methods, have been proposed and shown strong reasoning ability. Symbolic-based reasoning methods infer missing triples according to predefined rules or ontologies. Although rules and axioms have proven effective, it is difficult to obtain them. Embedding-based reasoning methods represent entities and relations as vectors, and complete KGs via vector computation. However, they mainly rely on structural information and ignore implicit axiom information not predefined in KGs but can be reflected in data. That is, each correct triple is also a logically consistent triple and satisfies all axioms. In this paper, we propose a novel NeuRal Axiom Network (NeuRAN) framework that combines explicit structural and implicit axiom information without introducing additional ontologies. Specifically, the framework consists of a KG embedding module that preserves the semantics of triples and five axiom modules that encode five kinds of implicit axioms. These axioms correspond to five typical object property expression axioms defined in OWL2, including ObjectPropertyDomain, ObjectPropertyRange, DisjointObjectProperties, IrreflexiveObjectProperty and AsymmetricObjectProperty. The KG embedding module and axiom modules compute the scores that the triple conforms to the semantics and the corresponding axioms, respectively. Compared with KG embedding models and CKRL, our method achieves comparable performance on noise detection and triple classification and achieves significant performance on link prediction. Compared with TransE and TransH, our method improves the link prediction performance on the Hits@1 metric by 22.0% and 20.8% on WN18RR-10% dataset, respectively.

Neural Models for Reasoning over Multiple Mentions Using Coreference

Coreferential Reasoning Learning for Language Representation

Improving Coreference Resolution by Leveraging Entity-Centric Features with Graph Neural Networks and Second-order Inference

Relational recurrent neural networks

Linguistic Knowledge as Memory for Recurrent Neural Networks.

Towards Neural Network-based Reasoning

Incorporating Singletons and Mention-based Features in Coreference Resolution via Multi-task Learning for Better Generalization

Neural Module Networks for Reasoning over Text

Neural Axiom Network for Knowledge Graph Reasoning

ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks

AdaLoGN: Adaptive Logic Graph Network for Reasoning-Based Machine Reading Comprehension

Modeling Hierarchical Reasoning Chains by Linking Discourse Units and Key Phrases for Reading Comprehension

A Bi-LSTM Mention Hypergraph Model with Encoding Schema for Mention Extraction

Verb-driven machine reading comprehension with dual-graph neural network

Online Coreference Resolution for Dialogue Processing: Improving Mention-Linking on Real-Time Conversations

Joint Learning for Coreference Resolution with Markov Logic

Parallel Data Helps Neural Entity Coreference Resolution

Okay, Let's Do This! Modeling Event Coreference with Generated Rationales and Knowledge Distillation

Coreference Resolution System for Indonesian Text with Mention Pair Method and Singleton Exclusion using Convolutional Neural Network

Coreference Resolution Based on High-Dimensional Multi-Scale Information

Bridging Context Gaps: Leveraging Coreference Resolution for Long Contextual Understanding