OpenCodePapers

question-answering-on-natural-questions-long

Question Answering
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeF1EMModelNameReleaseDate
Learning Dense Representations of Phrases at Scale✓ Link79.671.9DensePhrases2020-12-23
Cluster-Former: Clustering-based Sparse Transformer for Long-Range Dependency Encoding76.5Cluster-Former (#C=512)2020-09-13
Reformer: The Efficient Transformer✓ Link75.5Locality-Sensitive Hashing2020-01-13
Generating Long Sequences with Sparse Transformers✓ Link74.5Sparse Attention2019-04-23
Frustratingly Easy Natural Question Answering68.2BERTwwm + SQuAD 22019-09-11
A BERT Baseline for the Natural Questions✓ Link64.7BERTjoint2019-01-24
Natural Questions: a Benchmark for Question Answering Research✓ Link54.8DecAtt + DocReader2019-06-01
Reading Wikipedia to Answer Open-Domain Questions✓ Link46.1DrQA2017-03-31
0.8% Nyquist computational ghost imaging via non-experimental deep learning58.4FiE2021-08-17
R2-D2: A Modular Baseline for Open-Domain Question Answering✓ Link55.9R2-D2 \w HN-DPR2021-09-08
UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering✓ Link54.9UniK-QA2020-12-29
UnitedQA: A Hybrid Approach for Open Domain Question Answering54.7UnitedQA (Hybrid)2021-01-01
Efficient Passage Retrieval with Hashing for Open-domain Question Answering✓ Link41.6BPR (linear scan; l=1000)2021-06-02
Check out how to contribute new results.