ML Research Wiki / Benchmarks / Question Answering / NarrativeQA

NarrativeQA

Question Answering Benchmark

Performance Over Time

📊 Showing 8 results | 📏 Metric: Rouge-L

Top Performing Models

Rank Model Paper Rouge-L Date Code
1 BERT-QA with Hard EM objective A Discrete Hard EM Approach for Weakly Supervised Question Answering 58.80 2019-09-11 📦 shmsw25/qa-hard-em
2 Masque (NarrativeQA + MS MARCO) 📚 Multi-style Generative Reading Comprehension 30.43 2019-01-08 -
3 DecaProp Densely Connected Attention Propagation for Reading Comprehension 27.61 2018-11-10 📦 vanzytay/NIPS2018_DECAPROP 📦 ajenningsfrankston/NIPS2018_DECAPROP-master
4 MHPGM + NOIC Commonsense for Generative Multi-Hop Question Answering Tasks 21.07 2018-09-17 📦 yicheng-w/CommonSenseMultiHopQA 📦 a414351664/NarrativeQA
5 Masque (NarrativeQA only) Multi-style Generative Reading Comprehension 20.98 2019-01-08 -
6 BiDAF Bidirectional Attention Flow for Machine Comprehension 15.69 2016-11-05 📦 allenai/bi-att-flow 📦 baidu/DuReader 📦 galsang/BiDAF-pytorch
7 FiD+Distil Distilling Knowledge from Reader to Retriever for Question Answering 7.50 2020-12-08 📦 facebookresearch/FiD 📦 lucidrains/marge-pytorch 📦 hackerchenzhuo/LaKo 📦 FenQQQ/Fusion-in-decoder
8 Oracle IR Models The NarrativeQA Reading Comprehension Challenge 0.00 2017-12-19 📦 google-deepmind/narrativeqa 📦 deepmind/narrativeqa

All Papers (8)

shmsw25/qa-hard-em

Multi-style Generative Reading Comprehension

2019
Masque (NarrativeQA + MS MARCO)
vanzytay/NIPS2018_DECAPROP ajenningsfrankston/NIPS2018_DECAPROP-master
yicheng-w/CommonSenseMultiHopQA a414351664/NarrativeQA

Multi-style Generative Reading Comprehension

2019
Masque (NarrativeQA only)
allenai/bi-att-flow baidu/DuReader
facebookresearch/FiD lucidrains/marge-pytorch
google-deepmind/narrativeqa deepmind/narrativeqa