id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_wgi74n6oavhdxhlomowlxqjdhu	Oscar Täckström	Token and Type Constraints for Cross-Lingual Part-of-Speech Tagging	2013	12	.pdf	application/pdf	7667	915	74	Token and Type Constraints for Cross-Lingual Part-of-Speech Tagging where coupled token and type constraints provide a partial signal for training. Supervised part-of-speech (POS) taggers are available for more than twenty languages and achieve accuracies of around 95% on in-domain data (Petrov et POS taggers with type-level tag dictionary constraints Tag dictionaries, noisily projected via word-aligned bitext, have bridged the gap (CRF) model (Lafferty et al., 2001) that couples token and type constraints in order to guide learning type constraints and how we use these coupled constraints to train probabilistic tagging models. (2012).6 Since we use indirect supervision via projected tags or Wiktionary, the model Table 2: Tagging accuracies for models with token constraints and coupled token and type constraints. models that use only projected token constraints the token-level tag projections, so that the dictionary First, word types that occur with more token constraints during training are generally tagged more	./cache/work_wgi74n6oavhdxhlomowlxqjdhu.pdf	./txt/work_wgi74n6oavhdxhlomowlxqjdhu.txt