Data Efficiency
otherentity
ID: dataefficiency
Related Papers
Finding needles in a haystack: Sampling Structurally-diverse Training Sets from Synthetic Data for Compositional Generalization
AbstractModern semantic parsers suffer from two principal limitations. First, training requires expensive collection of utterance-program pairs. Second, semantic parsers fail to generalize at test time to new compositions/structures that have …...
Not All Linearizations Are Equally Data-Hungry in Sequence Labeling Parsing
AbstractDifferent linearizations have been proposed to cast dependency parsing as sequence labeling and solve the task as: (i) a head selection problem, (ii) finding a representation of the token arcs …...