Huda Khayrallah
@HudaKhay
Machine Translation/#NLProc/ML Researcher at Microsoft.
Past: @UCBerkeley CS ugrad; @LiltHQ research intern; @jhuCLSP/@jhuCompSci PhD
ID:720477252
http://khayrallah.github.io/ 27-07-2012 16:46:31
838 Tweets
1,0K Followers
914 Following
π’ MASC-SLL Call for Papers is out! π’
π Are you a student passionate about Speech, Language, and Learning? π£οΈβ¨
π Present your research at the Mid-Atlantic Student Colloquium on Speech, Language, and Learning (MASC-SLL) at Johns Hopkins Johns Hopkins Engineering on May 3rd! π
Heya, the Microsoft Translator team (more specifically me and Huda Khayrallah) is hiring for a senior researcher role.
Apply here!
jobs.careers.microsoft.com/global/en/job/β¦
From Armita R. Manafzadeh, a fundamental advance in our field β a true rarity in any discipline. In making this advance, she gives us an insight into the life of extinct dinosaurs of a sort that would have seemed impossible even a few years ago. Yale University Yale Peabody Museum Yale UniversityFaculty
This paper is the culmination of a research journey I embarked on nearly a decade ago β one which has changed the way I think about joints. It all started with grocery store chickens and protractors at UC Museum of Paleo, when I hoped to reconstruct locomotion from fossil bones.
For centuries, paleontologists have had to rely on intuition re whether reconstructed joint poses 'look right.' Today in Nature Communications: we created a way to formalize & test that intuition, & discovered that articular surface interactions distinguish locomotor poses in dinosaurs π§΅
π’In an upcoming TACL paper, Sweta Agrawal evaluated a wide range of text simplification models by asking an important but overlooked question: how well do readers understand key elements of the original content after reading the simplified version?π
π’ Exciting News! π
Our latest paper (w/Marine Carpuat) is out: 'Do Text Simplification Systems Preserve Meaning? A Human Evaluation via Reading Comprehension.' π