LING 575 E: Topics in Computational Linguistics

ANALYZING NEURAL LANGUAGE MODELS TWO RECENT TRENDS IN NLP---THE APPLICATION OF DEEP NEURAL NETWORKS AND THE USE OF TRANSFER LEARNING--- HAVE RESULTED IN MANY MODELS THAT ACHIEVE HIGH PERFORMANCE ON IMPORTA TASKS BUT WHOSE BEHAVIOR ON THOSE TASKS IS DIFFICULT TO INTERPR IN THIS SEMINAR, WE WILL LOOK AT METHODS INSPIRED BY LINGUISTICS COGNITIVE SCIENCE FOR ANALYZING WHAT LARGE NEURAL LANGUAGE MODELS H IN FACT LEARNED: DIAGNOSTIC/PROBING CLASSIFIERS, ADVERSARIAL TEST SETS, AND ARTIFICI LANGUAGES, AMONG OTHERS. PARTICULAR ATTENTION WILL BE PAID TO PROBING THESE MODELS' _SEMANTIC_ KNOWLEDGE, WHICH HAS RECEIVED COMPARABLY LITTLE ATTENTION COMPARED TO THEIR SYNTACTIC KNOWLEDGE. STUDENTS WILL ACQUIRE RELEVANT SKILLS AND (IN SMA GROUPS) DESIGN AND EXECUTE A LINGUISTICALLY-INFORMED ANALYSIS EXPERIMENT, RESULTING IN A REPORT I THE FORM OF A PUBLISHABLE CONFERENC PAPER. ON CAMPUS SECTION FOR CLMS STUDENTS.