HyperAIHyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAIHyperAI超神経

Relation Extraction On Ace 2004

評価指標

Cross Sentence
NER Micro F1
RE+ Micro F1

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Cross Sentence
NER Micro F1
RE+ Micro F1
Paper TitleRepository
Multi-turn QANo83.649.4Entity-Relation Extraction as Multi-Turn Question Answering
PFNNo89.362.5A Partition Filter Network for Joint Entity and Relation Extraction
Ours: cross-sentence ALBYes90.362.2A Frustratingly Easy Approach for Entity and Relation Extraction
PL-MarkerYes90.466.5Packed Levitated Marker for Entity and Relation Extraction
Joint w/ GlobalNo79.745.3--
SPTreeNo81.848.4End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures
AttentionNo79.645.7Going out on a limb: Joint Extraction of Entity Mentions and Relations without Dependency Trees-
Table-SequenceNo88.659.6Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders
multi-head + ATNo81.6447.45Adversarial training for multi-context joint entity and relation extraction
multi-headNo81.1647.14Joint entity recognition and relation extraction as a multi-head selection problem
DyGIEYes87.4-A General Framework for Information Extraction using Dynamic Span Graphs
0 of 11 row(s) selected.
Relation Extraction On Ace 2004 | SOTA | HyperAI超神経