id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_v3azwvbzizhhznkwmikacudzji	Xiang Lisa Li	A Generative Model for Punctuation in Dependency Trees	2019	17	.pdf	application/pdf	11820	1275	63	A Generative Model for Punctuation in Dependency Trees Also, our reconstruction of a sentence's underlying punctuation lets us appropriately render the surface Across the 5 languages, our model predicts surface punctuation Treebank annotation guidelines for punctuation tend to adopt simple heuristics like "attach to the highest possible node that preserves projectivity" (Bies et al., more underlying punctuation tokens.3 The probability p✓(l, r | w) is given by a log-linear model punctuation model on a subset of the Universal Dependencies (UD) version 1.4 (Nivre et al., Recall that our model generates surface punctuation given an unpunctuated dependency tree. We evaluate the trained punctuation model by using it in the following three tasks. model supposes that the observed punctuated sentence x̄ would have arisen via the generative process (1). Surface punctuation marks are not directly attached to the syntax tree, but are generated from sequences of adjacent punctemes by a (stochastic) finite-state string	./cache/work_v3azwvbzizhhznkwmikacudzji.pdf	./txt/work_v3azwvbzizhhznkwmikacudzji.txt