An LLM fills masked constituency trees to create a synthetic domain treebank, and span-level contrastive pre-training on it gives a new SOTA average F1 of 88.52 on cross-domain parsing.
Fast Rule-Based Decoding: Revisiting Syntactic Rules in Neural Constituency Parsing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Most recent studies on neural constituency parsing focus on encoder structures, while few developments are devoted to decoders. Previous research has demonstrated that probabilistic statistical methods based on syntactic rules are particularly effective in constituency parsing, whereas syntactic rules are not used during the training of neural models in prior work probably due to their enormous computation requirements. In this paper, we first implement a fast CKY decoding procedure harnessing GPU acceleration, based on which we further derive a syntactic rule-based (rule-constrained) CKY decoding. In the experiments, our method obtains 95.89 and 92.52 F1 on the datasets of PTB and CTB respectively, which shows significant improvements compared with previous approaches. Besides, our parser achieves strong and competitive cross-domain performance in zero-shot settings.
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Contrastive Learning on LLM Back Generation Treebank for Cross-domain Constituency Parsing
An LLM fills masked constituency trees to create a synthetic domain treebank, and span-level contrastive pre-training on it gives a new SOTA average F1 of 88.52 on cross-domain parsing.