| Attachment | Size |
|---|---|
 LoReHLT_SE_CFP_V2.txt | 2.84 KB |
CALL FOR PAPERS
Special Issue on NLP in Low-Resource Languages
Developing capabilities to extract information from a new language has traditionally been a very concerted, slow, and costly process due to high dependency on large, manually annotated data resources. The DARPA LORELEI program seeks to advance technologies that are less dependent on large data resources and that can be quickly pivoted to new languages within a very short amount of time so that information from any language can be extracted in timely manner to provide situation awareness to emergent incidents.
A special evaluation workshop, LoReHLT, was started by NIST in 2016 to support research on low-resource language NLP. In this evaluation, there is no training data in the evaluation language. Participants receive training data in related languages, but need to bootstrap systems in the surprise evaluation language at the start of the evaluation. Methods for this include pivoting approaches and taking advantage of linguistic universals.
This special journal issue looks to document promising new techniques developed in the LORELEI program as well as ideas and methods developed in the wider research community that target information extraction from low resource languages with a special focus on techniques that work across many languages, are less dependent on large data resources, take advantage of language universal resources, pivot from existing language resources to new incident languages, or bootstrap training resources for short development cycle. Of special interest are in the following areas:
* Machine translation
* Entities, relations, events extraction
* Sentiment detection
* Summarization
* Identifying locations mentioned in text
* Processing multiple genres (news, social media, conversational text…)
IMPORTANT DATES:
May 31, 2017: Paper submission due
June 30, 2017: Notification of acceptance
July 26, 2017: Camera ready paper due
SUBMISSION GUIDELINES:
* Authors should follow the "Instructions for Authors" available on the MT Journal website:
o Go to http://www.springer.com/computer/artificial/journal/10590
o Click on “Instructions for authors” on the right
o Expand “Text” and you will see a Latex template
* Length of paper is determined by total of submissions received. We recommend around 15 pages.
* Papers should be submitted online directly on the MT journal's submission website: http://www.editorialmanager.com/coat/default.asp and select “Special Issue on NLP in Low Resource Languages”
EDITORIAL COMMITTEE:
Ian Soboroff (NIST)
Audrey Tong (NIST)
Heng Ji (RPI)
Kevin Knight (ISI)
Lane Schwartz (UIUC)
Timothy Miller (UIUC)
Chen-Tse Tsai (UIUC)
Stephen Mayhew (UIUC)
Chao-Hong Liu (ADAPT)
Haithem Afli (ADAPT)
Iacer Calixto (ADAPT)
Longyue Wang (DCU)
Alberto Poncelas (DCU)
Jason Duncan (MITRE)
Ralph Weischedel (BBN)
Marjorie Freedman (BBN)