HyperAI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
HyperAI超神経

Question Answering On Natural Questions Long

評価指標

EM

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
EM
Paper TitleRepository
FiE58.40.8% Nyquist computational ghost imaging via non-experimental deep learning-
DensePhrases71.9Learning Dense Representations of Phrases at Scale
R2-D2 w HN-DPR55.9R2-D2: A Modular Baseline for Open-Domain Question Answering-
UnitedQA (Hybrid)54.7UnitedQA: A Hybrid Approach for Open Domain Question Answering-
BERTwwm + SQuAD 2-Frustratingly Easy Natural Question Answering-
Cluster-Former (#C=512)-Cluster-Former: Clustering-based Sparse Transformer for Long-Range Dependency Encoding-
DrQA-Reading Wikipedia to Answer Open-Domain Questions
Locality-Sensitive Hashing-Reformer: The Efficient Transformer
UniK-QA54.9UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering
BERTjoint-A BERT Baseline for the Natural Questions
Sparse Attention-Generating Long Sequences with Sparse Transformers
BPR (linear scan; l=1000)41.6Efficient Passage Retrieval with Hashing for Open-domain Question Answering
DecAtt + DocReader-Natural Questions: a Benchmark for Question Answering Research
0 of 13 row(s) selected.