Variational Reasoning for Question Answering With Knowledge Graph

Zhang, Yuyu; Dai, Hanjun; Kozareva, Zornitsa; Smola, Alexander J.; Song, Le

doi:10.1609/aaai.v32i1.12057

Cited by 194 publications

(67 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the vanilla version of the MetaQA (Zhang et al 2018) dataset, which contains questions requiring multi-hop reasoning over a novel movie-domain knowledge graph. Each question is provided with one entity mention and the question is named as a k-hop question if the answer entity is a khop neighbor of the question entity.…”

Section: Downstream Tasksmentioning

confidence: 99%

JAKET: Joint Pre-training of Knowledge Graph and Language Understanding

Zhu

Yang

et al. 2022

AAAI

View full text Add to dashboard Cite

Knowledge graphs (KGs) contain rich information about world knowledge, entities, and relations. Thus, they can be great supplements to existing pre-trained language models. However, it remains a challenge to efficiently integrate information from KG into language modeling. And the understanding of a knowledge graph requires related context. We propose a novel joint pre-training framework, JAKET, to model both the knowledge graph and language. The knowledge module and language module provide essential information to mutually assist each other: the knowledge module produces embeddings for entities in text while the language module generates context-aware initial embeddings for entities and relations in the graph. Our design enables the pre-trained model to easily adapt to unseen knowledge graphs in new domains. Experiment results on several knowledge-aware NLP tasks show that our proposed framework achieves superior performance by effectively leveraging knowledge in language understanding.

show abstract

Section: Downstream Tasksmentioning

confidence: 99%

JAKET: Joint Pre-training of Knowledge Graph and Language Understanding

Zhu

Yang

et al. 2022

AAAI

View full text Add to dashboard Cite

show abstract

“…However, for answer retrieval, in contrast with embedding-based methods that utilize one-step operations (e.g., distance and similarity computations of embeddings), reasoning-based methods employ more complex sequential operations that can perform multi-step reasoning on KGs regarding given questions. Examples of this type of methods include [6,10,22,23,28,29]. One commonly followed paradigm is to convert question answering into a sequential decision-making problem and train a reinforcement learning-based QA agent.…”

Section: Related Workmentioning

confidence: 99%

“…Datasets We adopted two widely used benchmark datasets: PathQuestion [29] and MetaQA [28]. Dataset statistics can be found in Table 1.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…A topic entity Sleepy Hollow is indicated in the question, and the expected answer Jess Talamantes needs to be inferred from a path that starts from the topic entity and consists of three KG triples: {(Sleepy Hollow, director, Tim Burton), (Tim Burton, birthplace, Burbank), (Burbank, mayor, Jess Talamantes)}. Reasoning-based methods [6,22,28,29] have been the mainstream approach for answering multirelation questions. They can perform multi-step reasoning over KGs to explore paths that infer expected answers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

QAGCN: A Graph Convolutional Network-based Multi-Relation Question Answering System

Wang¹,

Rossetto²,

Cochez³

et al. 2022

Preprint

View full text Add to dashboard Cite

Answering multi-relation questions over knowledge graphs is a challenging task as it requires multi-step reasoning over a huge number of possible paths. Reasoning-based methods with complex reasoning mechanisms, such as reinforcement learning-based sequential decision making, have been regarded as the default pathway for this task. However, these mechanisms are difficult to implement and train, which hampers their reproducibility and transferability to new domains. In this paper, we propose QAGCN -a simple but effective and novel model that leverages attentional graph convolutional networks that can perform multi-step reasoning during the encoding of knowledge graphs. As a consequence, complex reasoning mechanisms are avoided. In addition, to improve efficiency, we retrieve answers using highly-efficient embedding computations and, for better interpretability, we extract interpretable paths for returned answers. On widely adopted benchmark datasets, the proposed model has been demonstrated competitive against state-of-theart methods that rely on complex reasoning mechanisms. We also conducted extensive experiments to scrutinize the efficiency and contribution of each component of our model.

show abstract

“…Variational Reasoning Network (VRN) VRN (Zhang et al, 2018) proposes a variational framework for multi-hop KGQA. To identify answers, it computes the compatibility scores between the question type and the reasoning graph of each candidate.…”

Section: Multi-hop Kbqamentioning

confidence: 99%

Deep Cognitive Reasoning Network for Multi-hop Question Answering over Knowledge Graphs

Cai¹,

Zhang²,

Wu³

et al. 2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

View full text Add to dashboard Cite

Knowledge Graphs (KGs) provide human knowledge with nodes and edges being entities and relations among them, respectively. Multihop question answering over KGs-which aims to find answer entities of given questions through reasoning paths in KGs-has attracted great attention from both academia and industry recently. However, this task remains challenging, as it requires to accurately identify answers in a large candidate entity set, of which the size grows exponentially with the number of reasoning hops. To tackle this problem, we propose a novel Deep Cognitive Reasoning Network (DCRN), which is inspired by the dual process theory in cognitive science. Specifically, DCRN consists of two phases-the unconscious phase and the conscious phase. The unconscious phase first retrieves informative evidence from candidate entities by leveraging their semantic information. Then, the conscious phase accurately identifies answers by performing sequential reasoning according to the graph structure on the retrieved evidence. Experiments demonstrate that DCRN significantly outperforms state-of-the-art methods on benchmark datasets.

show abstract

Variational Reasoning for Question Answering With Knowledge Graph

Cited by 194 publications

References 21 publications

JAKET: Joint Pre-training of Knowledge Graph and Language Understanding

JAKET: Joint Pre-training of Knowledge Graph and Language Understanding

QAGCN: A Graph Convolutional Network-based Multi-Relation Question Answering System

Deep Cognitive Reasoning Network for Multi-hop Question Answering over Knowledge Graphs

Contact Info

Product

Resources

About