OpenCodePapers
question-answering-on-narrativeqa
Question Answering
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Show papers without code
Paper
Code
Rouge-L
↕
BLEU-1
↕
BLEU-4
↕
METEOR
↕
ModelName
ReleaseDate
↕
Multi-style Generative Reading Comprehension
59.87
54.11
30.43
26.13
Masque (NarrativeQA + MS MARCO)
2019-01-08
A Discrete Hard EM Approach for Weakly Supervised Question Answering
✓ Link
58.8
BERT-QA with Hard EM objective
2019-09-11
Multi-style Generative Reading Comprehension
54.74
48.7
20.98
21.95
Masque (NarrativeQA only)
2019-01-08
Cut to the Chase: A Context Zoom-in Network for Reading Comprehension
46.67
42.76
22.49
19.24
ConZNet
2018-10-01
Densely Connected Attention Propagation for Reading Comprehension
✓ Link
44.69
44.35
27.61
21.80
DecaProp
2018-11-10
Commonsense for Generative Multi-Hop Question Answering Tasks
✓ Link
44.16
43.63
21.07
19.03
MHPGM + NOIC
2018-09-17
Multi-Granular Sequence Encoding via Dilated Compositional Units for Reading Comprehension
41.44
36.55
19.79
17.87
BiAttention + DCU-LSTM
2018-10-01
Bidirectional Attention Flow for Machine Comprehension
✓ Link
36.74
33.45
15.69
15.68
BiDAF
2016-11-05
Distilling Knowledge from Reader to Retriever for Question Answering
✓ Link
32
35.3
7.5
11.1
FiD+Distil
2020-12-08
The NarrativeQA Reading Comprehension Challenge
✓ Link
54.60/55.55
26.71/27.78
Oracle IR Models
2017-12-19