Share this page:

A Paragraph-level Multi-task Learning Model for Scientific Fact-Verification

Xiangci Li, Gully Burns, and Nanyun Peng, in Scientific Document Understanding Workshop at the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21), 2021.

Download the full text


Abstract

Even for domain experts, it is a non-trivial task to verify a scientific claim by providing supporting or refuting evidence rationales. The situation worsens as misinformation is proliferated on social media or news websites, manually or programmatically, at every moment. As a result, an automatic fact-verification tool becomes crucial for combating the spread of misinformation. In this work, we propose a novel, paragraph-level, multi-task learning model for the SciFact task by directly computing a sequence of contextualized sentence embeddings from a BERT model and jointly training the model on rationale selection and stance prediction.


Bib Entry

@inproceedings{li2021paragraph,
  title = {A Paragraph-level Multi-task Learning Model for Scientific Fact-Verification},
  author = {Li, Xiangci and Burns, Gully and Peng, Nanyun},
  booktitle = {Scientific Document Understanding Workshop at the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21)},
  year = {2021}
}

Related Publications

  • A Paragraph-level Multi-task Learning Model for Scientific Fact-Verification

    Xiangci Li, Gully Burns, and Nanyun Peng, in Scientific Document Understanding Workshop at the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21), 2021.
    Full Text Code Abstract BibTeX Details
    Even for domain experts, it is a non-trivial task to verify a scientific claim by providing supporting or refuting evidence rationales. The situation worsens as misinformation is proliferated on social media or news websites, manually or programmatically, at every moment. As a result, an automatic fact-verification tool becomes crucial for combating the spread of misinformation. In this work, we propose a novel, paragraph-level, multi-task learning model for the SciFact task by directly computing a sequence of contextualized sentence embeddings from a BERT model and jointly training the model on rationale selection and stance prediction.
    @inproceedings{li2021paragraph,
      title = {A Paragraph-level Multi-task Learning Model for Scientific Fact-Verification},
      author = {Li, Xiangci and Burns, Gully and Peng, Nanyun},
      booktitle = {Scientific Document Understanding Workshop at the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21)},
      year = {2021}
    }
    
    Details