On December 22, the Intelligence Advanced Research Projects Activity, within the Office of the Director of National Intelligence, announced a multi-year research effort to develop language processing software, known as Machine Translation for English Retrieval of Information in Any Language—“MATERIAL”— that will enable users to quickly develop and deploy fully automatic systems that will allow English-only speakers to accurately and efficiently identify foreign language documents of interest across social media, newswire, and broadcasts—to name a few.
“The collection and analysis of information required to accomplish a specific intelligence task has increasingly become a multilingual venture,” said Dr. Carl Rubino, IARPA program manager. “For most languages, there are very few or no automated tools available for cross-lingual data mining and analysis. The MATERIAL Program aims to investigate how current language processing technologies can most efficiently be developed and integrated to respond to specific information needs against multilingual speech and text data.”
The MATERIAL Program intends to demonstrate the ability to quickly develop cross-language information retrieval systems, in language-independent ways that can be easily adapted to new domains of interest. Program participants will also develop cross-language summarization technologies to enable more efficient processing of the data.
Through a competitive Broad Agency Announcement, IARPA has awarded MATERIAL’s research contracts to teams led by Johns Hopkins University, Raytheon BBN Technologies, Columbia University and University of Southern California Information Sciences Institute. MIT Lincoln Laboratory, University of Maryland Center for Advanced Study of Language, National Institute of Standards and Technology and Tarragon Consulting comprise the MATERIAL Test and Evaluation team which will assess performance of a variety of complex End-to-End solutions developed by the performers.