COMET: Commonsense Transformers for Automatic Knowledge Graph Construction

@inproceedings{Bosselut2019COMETCT,
  title={COMET: Commonsense Transformers for Automatic Knowledge Graph Construction},
  author={Antoine Bosselut and Hannah Rashkin and Maarten Sap and Chaitanya Malaviya and A. Çelikyilmaz and Yejin Choi},
  booktitle={ACL},
  year={2019}
}
  • Antoine Bosselut, Hannah Rashkin, +3 authors Yejin Choi
  • Published in ACL 2019
  • Computer Science
  • We present the first comprehensive study on automatic knowledge base construction for two prevalent commonsense knowledge graphs: ATOMIC (Sap et al., 2019) and ConceptNet (Speer et al., 2017. [...] Key Result Empirical results demonstrate that COMET is able to generate novel knowledge that humans rate as high quality, with up to 77.5% (ATOMIC) and 91.7% (ConceptNet) precision at top 1, which approaches human performance for these resources.Expand Abstract
    110 Citations
    Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering
    • 13
    Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering
    COMET-ATOMIC 2020: On Symbolic and Neural Commonsense Knowledge Graphs
    Exploiting Structural and Semantic Context for Commonsense Knowledge Base Completion
    • 8
    Commonsense Knowledge Base Completion with Structural and Semantic Context
    • 9
    • PDF
    Connecting the Dots: A Knowledgeable Path Generator for Commonsense Question Answering
    • 2
    • Highly Influenced
    • PDF
    On the Role of Conceptualization in Commonsense Knowledge Graph Construction
    • 1
    • Highly Influenced
    • PDF
    Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graph
    • 2
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 37 REFERENCES
    Commonsense Knowledge Base Completion
    • 71
    • Highly Influential
    • PDF
    Commonsense Knowledge Base Completion and Generation
    • 16
    • Highly Influential
    • PDF
    ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning
    • 138
    • PDF
    Knowledge vault: a web-scale approach to probabilistic knowledge fusion
    • 1,065
    • Highly Influential
    • PDF
    ConceptNet 5.5: An Open Multilingual Graph of General Knowledge
    • 571
    • PDF
    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
    • 11,729
    • PDF
    Yago: a core of semantic knowledge
    • 2,997
    • Highly Influential
    • PDF
    Scalable knowledge harvesting with high precision and high recall
    • 189
    • PDF
    Zero-Shot Relation Extraction via Reading Comprehension
    • 138
    • PDF
    Web-scale knowledge-base construction via statistical inference and learning
    • 2