The UD-NewsCrawl Treebank: Reflections and Challenges from a Large-scale Tagalog Syntactic Annotation Project
			Paper
			•
			2505.20428
			•
			Published
				
			
			
Models and dependency parsers for Tagalog using the UD_NewsCrawl dataset
 
				Note spaCy pipeline using a transition-based parser (baseline)
 
				Note spaCy pipeline using context-sensitive vectors from XLM-RoBERTa and a transition-based parser.
 
				Note spaCy pipeline using context-sensitive vectors from RoBERTa-Tagalog and a transition-based parser
 
				Note spaCy pipeline using context-sensitive vecotrs from mDeBERTa-v3 and a transition-based parser
 
				Note spaCy pipeline using fastText word embeddings and a transition-based parser
 
				Note spaCy pipeline using multi hash embeddings and a transition-based parser