Papers › HUJI-KU at MRP~2020: Two Transition-based Neural Parsers

HUJI-KU at MRP~2020: Two Transition-based Neural Parsers

12 Oct 2020arXiv:2010.05710archive 2025-07-28

Ofir Arviv, Ruixiang Cui, Daniel Hershcovich

This paper describes the HUJI-KU system submission to the shared task on Cross-Framework Meaning Representation Parsing (MRP) at the 2020 Conference for Computational Language Learning (CoNLL), employing TUPA and the HIT-SCIR parser, which were, respectively, the baseline system and winning system in the 2019 MRP shared task. Both are transition-based parsers using BERT contextualized embeddings. We generalized TUPA to support the newly-added MRP frameworks and languages, and experimented with multitask learning with the HIT-SCIR parser. We reached 4th place in both the cross-framework and cross-lingual tracks.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Semantic ParsingVocal Bursts Valence Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semantic Parsing AMR (chinese, MRP 2020) HUJI-KU F1 45 #2 of 2 Archive leaderboard report
Semantic Parsing AMR (english, MRP 2020) HUJI-KU F1 52 #2 of 2 Archive leaderboard report
Semantic Parsing DRG (english, MRP 2020) HUJI-KU F1 63 #2 of 2 Archive leaderboard report
Semantic Parsing DRG (german, MRP 2020) HUJI-KU F1 62 #2 of 2 Archive leaderboard report
Semantic Parsing EDS (english, MRP 2020) HUJI-KU F1 80 #2 of 2 Archive leaderboard report
Semantic Parsing PTG (czech, MRP 2020) HUJI-KU F1 58 #3 of 3 Archive leaderboard report
Semantic Parsing PTG (english, MRP 2020) HUJI-KU F1 54 #2 of 2 Archive leaderboard report
Semantic Parsing UCCA (english, MRP 2020) HUJI-KU F1 73 #2 of 2 Archive leaderboard report
Semantic Parsing UCCA (german, MRP 2020) HUJI-KU F1 75 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionSoftmaxWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections