Skip to content

Version 0.4.3 of GROBID

Compare
Choose a tag to compare
@kermitt2 kermitt2 released this 07 Oct 00:55
· 2398 commits to master since this release
grobid-parent-0.4.3
89c0833

The latest stable release of GROBID is version 0.4.3. As compared to previous version 0.4.2, this version brings:

  • New models: f-score improvement on the PubMed Central sample, bibliographical references +2.5%, header +7%
  • New training data and features for bibliographical references, in particular for covering HEP domain (INSPIRE), arXiv identifier, DOI and url (thanks @iorala and @michamos !)
  • Support for CrossRef REST API (instead of the slow OpenURL-style API which requires a CrossRef account), in particular for multithreading usage (thanks @Vi-dot)
  • Improve training data generation and documentation (thanks @jfix)
  • Unicode normalisation and more robust body extraction (thanks @aoboturov)
  • fixes, tests, documentation and update of the pdf2xml fork for Windows (thanks @lfoppiano)
Assets 2
Loading