ÚFAL at MRP 2020: Permutation-invariant Semantic Parsing in PERIN
2020-11-02Code Available1· sign in to hype
David Samuel, Milan Straka
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/ufal/perinOfficialIn paperpytorch★ 45
- github.com/gotheregit/chinese-amrnone★ 39
Abstract
We present PERIN, a novel permutation-invariant approach to sentence-to-graph semantic parsing. PERIN is a versatile, cross-framework and language independent architecture for universal modeling of semantic structures. Our system participated in the CoNLL 2020 shared task, Cross-Framework Meaning Representation Parsing (MRP 2020), where it was evaluated on five different frameworks (AMR, DRG, EDS, PTG and UCCA) across four languages. PERIN was one of the winners of the shared task. The source code and pretrained models are available at https://github.com/ufal/perin.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| AMR (chinese, MRP 2020) | PERIN | F1 | 80.52 | — | Unverified |
| AMR (english, MRP 2020) | PERIN | F1 | 80.23 | — | Unverified |
| DRG (english, MRP 2020) | PERIN | F1 | 94.16 | — | Unverified |
| DRG (german, MRP 2020) | PERIN | F1 | 89.83 | — | Unverified |
| EDS (english, MRP 2020) | PERIN | F1 | 92.73 | — | Unverified |
| PTG (czech, MRP 2020) | PERIN | F1 | 92.24 | — | Unverified |
| PTG (english, MRP 2020) | PERIN | F1 | 89.19 | — | Unverified |
| UCCA (english, MRP 2020) | PERIN | F1 | 76.4 | — | Unverified |
| UCCA (german, MRP 2020) | PERIN | F1 | 81.01 | — | Unverified |