SOTAVerified

AMR Parsing

Each AMR is a single rooted, directed graph. AMRs include PropBank semantic roles, within-sentence coreference, named entities and types, modality, negation, questions, quantities, and so on. See.

Papers

Showing 61–70 of 117 papers

TitleStatusHype
A Differentiable Relaxation of Graph Segmentation and Alignment for AMR Parsing—0
Transition-based Parsing with Stack-Transformers—0
Pushing the Limits of AMR Parsing with Self-Learning—0
AMR Parsing with Latent Structural Information—0
Constructing Web-Accessible Semantic Role Labels and Frames for Japanese as Additions to the NPCMJ Parsed Corpus—0
Core Semantic First: A Top-down Approach for AMR ParsingCode0
Broad-Coverage Semantic Parsing as Transduction—0
Exploring Graph-Algebraic CCG Combinators for Syntactic-Semantic AMR Parsing—0
Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning—0
Automatic Accuracy Prediction for AMR Parsing—0
Show:102550
← PrevPage 7 of 12Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1StructBART + MBSE (IBM)Smatch86.7—Unverified
2Graphene Smatch (IBM)Smatch86.26—Unverified
3LeakDistillSmatch86.1—Unverified
4StructBART + MBSE (IBM)Smatch85.9—Unverified
5Graphene Support (IBM)Smatch85.85—Unverified
6AMRBART largeSmatch85.4—Unverified
7ATP-SRL (Ensemble)Smatch85.3—Unverified
8ATP-SRLSmatch85.2—Unverified
9StructBART-J (IBM)Smatch84.7—Unverified
10BiBL+SilverSmatch84.7—Unverified
#ModelMetricClaimedVerifiedStatus
1APT (IBM)F1 Full78.5—Unverified
2stack-Transformer + self-learning (IBM)F1 Full78.2—Unverified
3AMR Parsing via Graph-Sequence Iterative InferenceF1 Full75.4—Unverified
4Broad-Coverage Semantic Parsing as TransductionF1 Full71.3—Unverified
5Two-stage Sequence-to-Graph TransducerF1 Full70.2—Unverified
6Imitation learningF1 Newswire70—Unverified
7Transition-based+improved aligner+ensembleF1 Full68.4—Unverified
8Improved CAMR F1 Full68.1—Unverified
9Transition-based transducerF1 Full66—Unverified
10Incremental joint model F1 Full66—Unverified
#ModelMetricClaimedVerifiedStatus
1Graphene Smatch (MBSE paper) (IBM)Smatch85.4—Unverified
2Graphene Smatch (IBM)Smatch84.87—Unverified
3LeakDistillSmatch84.6—Unverified
4Graphene Support (IBM)Smatch84.41—Unverified
5StructBART + MBSE (IBM)Smatch84.3—Unverified
6AMRBART largeSmatch84.2—Unverified
7ATP-SRLSmatch84—Unverified
8BiBLSmatch83.9—Unverified
9BiBL+SilverSmatch83.5—Unverified
10LeakDistill (base)Smatch83.5—Unverified
#ModelMetricClaimedVerifiedStatus
1Joint modelSmatch73.7—Unverified
2Mul-BiLSTMSmatch70.7—Unverified
3JAMRSmatch67—Unverified
4CAMRSmatch66.5—Unverified
5AMREagerSmatch64—Unverified
6SEQ2SEQ + 20MSmatch62.1—Unverified
#ModelMetricClaimedVerifiedStatus
1StructBART + MBSE (IBM)Smatch66.9—Unverified
2AMRBART largeSmatch63.2—Unverified
3Graphene SmatchSmatch62.8—Unverified
4SPRING DFSSmatch59.7—Unverified
5SPRING DFS + silverSmatch59.5—Unverified
#ModelMetricClaimedVerifiedStatus
1AMRBART largeSmatch76.9—Unverified
2Graphene SmatchSmatch76.32—Unverified
3SPRING DFSSmatch73.7—Unverified
4SPRING DFS + silverSmatch71.8—Unverified
#ModelMetricClaimedVerifiedStatus
1AMRBART largeSmatch79.8—Unverified
2Graphene SmatchSmatch79.52—Unverified
3SPRING DFS + silverSmatch77.5—Unverified
4SPRING DFSSmatch77.3—Unverified