A new BibTeX-normalizing tool dubbed Rebiber is gaining popularity in the AI research community. The creation of a PhD student, Rebiber addresses incomplete or confusing paper citation information.
BibTex is a popular reference management software that can automate most of the work involved in managing references for use in LaTeX files. It is commonly used by researchers for formatting lists of references when citing other papers. Users first create a bibliography database and store it as a plain text .bib file that can be easily viewed and edited.
Rebiber creator and University of Southern California Computer Science PhD student (Bill) Yuchen Lin however noticed that many AI researchers only cite paper information from the popular preprint arXiv, and do not reference information associated with conferences such as ACL, EMNLP, NAACL, ICLR, or AAAI. “These incorrect bib entries might violate rules about submissions or camera-ready versions for conferences,” Lin explains.
Lin designed the simple Python tool to automatically update .bib entries with references to papers’ official conference information.
Lin collected the conference information for the references from the full ACL anthology and DBLP (The dblp computer science bibliography), and compiled them into the .bib data.
Rebiber can be accessed for free on Lin’s GitHub repository. It includes all papers and workshop documentation published at computational linguistic and natural language processing conferences such as ACL EMNLP, NAACL, etc. It also supports any conference proceedings downloaded from DBLP, such as ICLR2020, NeurIPS 2020, etc.
Reporter: Fangyu Cai | Editor: Michael Sarazen
good article