Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims
This paper presents the latest developments to ClaimBuster?s claim-spotting model, which tackles the critical task of identifying check-worthy claims from large streams of information. We introduce the first adversarially-regularized, transformer-based claim-spotting model, which achieves state-of-t...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
ACM
2024
|
| Online Access: | https://hdl.handle.net/1721.1/156666 |
| _version_ | 1826197538157363200 |
|---|---|
| author | Meng, Kevin Jimenez, Damian Devasier, Jacob Naraparaju, Sai Sandeep Arslan, Fatma Obembe, Daniel Li, Chengkai |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Meng, Kevin Jimenez, Damian Devasier, Jacob Naraparaju, Sai Sandeep Arslan, Fatma Obembe, Daniel Li, Chengkai |
| author_sort | Meng, Kevin |
| collection | MIT |
| description | This paper presents the latest developments to ClaimBuster?s claim-spotting model, which tackles the critical task of identifying check-worthy claims from large streams of information. We introduce the first adversarially-regularized, transformer-based claim-spotting model, which achieves state-of-the-art results on several bench-mark datasets. In addition to analyzing model performance metrics, we also quantitatively and qualitatively analyze the impact of ClaimBuster?s real-world deployment. Moreover, to help facilitate reproducibility and community engagement, we publicly release our codebase, dataset, data curation platform, API, Google Colab notebooks, and various ClaimBuster-based demo systems, at claimbuster.org. |
| first_indexed | 2024-09-23T10:49:12Z |
| format | Article |
| id | mit-1721.1/156666 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2025-02-19T04:19:38Z |
| publishDate | 2024 |
| publisher | ACM |
| record_format | dspace |
| spelling | mit-1721.1/1566662024-12-21T06:01:40Z Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims Meng, Kevin Jimenez, Damian Devasier, Jacob Naraparaju, Sai Sandeep Arslan, Fatma Obembe, Daniel Li, Chengkai Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory This paper presents the latest developments to ClaimBuster?s claim-spotting model, which tackles the critical task of identifying check-worthy claims from large streams of information. We introduce the first adversarially-regularized, transformer-based claim-spotting model, which achieves state-of-the-art results on several bench-mark datasets. In addition to analyzing model performance metrics, we also quantitatively and qualitatively analyze the impact of ClaimBuster?s real-world deployment. Moreover, to help facilitate reproducibility and community engagement, we publicly release our codebase, dataset, data curation platform, API, Google Colab notebooks, and various ClaimBuster-based demo systems, at claimbuster.org. 2024-09-04T17:37:54Z 2024-09-04T17:37:54Z 2024-09-01T07:45:32Z Article http://purl.org/eprint/type/JournalArticle 2157-6904 https://hdl.handle.net/1721.1/156666 Kevin Meng, Damian Jimenez, Jacob Daniel Devasier, Sai Sandeep Naraparaju, Fatma Arslan, Daniel Obembe, and Chengkai Li. 2024. Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims. ACM Trans. Intell. Syst. Technol. Just Accepted (August 2024). PUBLISHER_POLICY en 10.1145/3689212 ACM Transactions on Intelligent Systems and Technology Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. The author(s) application/pdf ACM Association for Computing Machinery |
| spellingShingle | Meng, Kevin Jimenez, Damian Devasier, Jacob Naraparaju, Sai Sandeep Arslan, Fatma Obembe, Daniel Li, Chengkai Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title | Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title_full | Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title_fullStr | Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title_full_unstemmed | Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title_short | Gradient-Based Adversarial Training on Transformer Networks for Detecting Check-Worthy Factual Claims |
| title_sort | gradient based adversarial training on transformer networks for detecting check worthy factual claims |
| url | https://hdl.handle.net/1721.1/156666 |
| work_keys_str_mv | AT mengkevin gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT jimenezdamian gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT devasierjacob gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT naraparajusaisandeep gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT arslanfatma gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT obembedaniel gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims AT lichengkai gradientbasedadversarialtrainingontransformernetworksfordetectingcheckworthyfactualclaims |