TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Natural Language Processing/Question Answering/SQuAD1.1

Question Answering on SQuAD1.1

Metric: F1 (higher is better)

LeaderboardDataset
Loading chart...

Results

Submit a result
#Model↕F1▼Extra DataPaperDate↕Code
1{ANNA} (single model)95.719Yes---
2LUKE 483M95.4NoLUKE: Deep Contextualized Entity Representations...2020-10-02Code
3LUKE (single model)95.379YesLUKE: Deep Contextualized Entity Representations...2020-10-02Code
4LUKE (single model)95.379NoLUKE: Deep Contextualized Entity Representations...2020-10-02Code
5XLNet (single model)95.08NoXLNet: Generalized Autoregressive Pretraining fo...2019-06-19Code
6XLNet (single model)95.08YesXLNet: Generalized Autoregressive Pretraining fo...2019-06-19Code
7XLNET-123 (single model)94.93No---
8XLNET-123++ (single model)94.903No---
9XLNET-123+ (single model)94.859No---
10SpanBERT (single model)94.635No---
11SpanBERT (single model)94.6NoSpanBERT: Improving Pre-training by Representing...2019-07-24Code
12Unnamed submission by NMC94.584No---
13BERTSP (single model)94.584No---
14BERT+WWM+MT (single model)94.393No---
15Tuned BERT-1seq Large Cased (single model)93.294No---
16BERT-LARGE (Ensemble+TriviaQA)93.2NoBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
17BERT (ensemble)93.16NoBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
18BART (TextBox 2.0)93.04NoTextBox 2.0: A Text Generation Library with Pre-...2022-12-26Code
19LinkBERT (large)92.7NoLinkBERT: Pretraining Language Models with Docum...2022-03-29Code
20BERT+MT (single model)92.645No---
21ATB (single model)92.641No---
22Tuned BERT Large Cased (single model)92.617No---
23Knowledge-enhanced BERT (single model)92.425No---
24KT-NET (single model)92.425No---
25DPN (single model)92.019No---
26ST_bl91.976No---
27BERT-uncased (single model)91.932No---
28BERT (single model)91.835YesBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
29EL-BERT (single model)91.807No---
30BERT-LARGE (Single+TriviaQA)91.8NoBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
31BISAN (single model)91.756No---
32BERT+Sparse-Transformer91.623No---
33BERT-Large 32k batch size with AdamW91.58NoA Large Batch Optimizer Reality Check: Tradition...2021-02-12-
34Original BERT Large Cased (single model)91.281No---
35nlnet (ensemble)91.202No---
36DyREX91.01NoDyREx: Dynamic Query Representation for Extracti...2022-10-26Code
37Common-sense Governed BERT-123 (single model)90.613No---
38WD (single model)90.561No---
39WD1 (single model)90.429No---
40nlnet (single model)90.133No---
41MARS (ensemble)89.796No---
42BERT-Base mod (single model)89.379No---
43QANet (single)89.306No---
44Hybrid AoA Reader (ensemble)89.281No---
45Pytalk + Stanza + BERT (single model)89.218No---
46MMIPN88.948No---
47BERT (single model)88.947No---
48ARSG-BERT (single model)88.909No---
49Reinforced Mnemonic Reader + A2D (ensemble model)88.764No---
50SLQA+ (ensemble)88.607No---
51Reinforced Mnemonic Reader (ensemble model)88.533YesReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
52BERT - 6 Layers88.5NoInformation Theoretic Representation Distillation2021-12-01Code
53r-net+ (ensemble)88.493No---
54batch (single model)88.263No---
55mBERT + Task Adapter (Single)88.169No---
56AttentionReader+ (ensemble)88.163No---
57r-net (ensemble)88.126No---
58Reinforced Mnemonic Reader + A2D + DA (single model)88.122No---
59BERT-COMPOUND-DSS (single model)87.999No---
60BERT-COMPOUND (single model)87.758No---
61KACTEIL-MRC(GF-Net+) (ensemble)87.557No---
62Reinforced Mnemonic Reader + A2D (single model)87.454No---
63BiDAF + Self Attention + ELMo (ensemble)87.432NoDeep contextualized word representations2018-02-15Code
64BiDAF + Self Attention + ELMo (ensemble)87.432NoDeep contextualized word representations2018-02-15-
65BERT-INDEPENDENT-DSS-FILTERED (single model)87.374No---
66AVIQA+ (ensemble)87.311No---
67Hybrid AoA Reader (single model)87.288No---
68SLQA+87.021No---
69{EAZI} (ensemble)86.912No---
70EAZI+ (ensemble)86.912No---
71MAMCN+ (single model)86.727No---
72MAMCN+ (single model)86.727No---
73DNET (ensemble)86.721No---
74BiDAF + Self Attention + ELMo + A2D (single model)86.711No---
75BERT-INDEPENDENT (single model)86.663No---
76Reinforced Mnemonic Reader (single model)86.654NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
77SLQA+ (single model)86.59No---
78r-net+ (single model)86.536No---
79SAN (ensemble model)86.496NoStochastic Answer Networks for Machine Reading C...2017-12-10Code
80Interactive AoA Reader+ (ensemble)86.45No---
81MIR-MRC(F-Net) (single model)86.288No---
82KACTEIL-MRC(GF-Net+Distillation) (single model)86.288No---
83KACTEIL-MRC (GF-Net+Distillation)86.288No---
84FusionNet (ensemble)86.016NoFusionNet: Fusing via Fully-Aware Attention with...2017-11-16Code
85MDReader86.006No---
86DCN+ (ensemble)85.996NoDCN+: Mixed Objective and Deep Residual Coattent...2017-10-31Code
87BiDAF + Self Attention + ELMo (single model)85.833NoDeep contextualized word representations2018-02-15Code
88BiDAF + Self Attention + ELMo (single model)85.833NoDeep contextualized word representations2018-02-15-
89BERT - 3 Layers85.8NoInformation Theoretic Representation Distillation2021-12-01Code
90KACTEIL-MRC(GF-Net+) (single model)85.78No---
91KACTEIL-MRC (GF-Net+)85.78No---
92KakaoNet (single model)85.724No---
93SLQA(ensemble)85.682No---
94SLQA (ensemble)85.682No---
95MDReader085.543No---
96BiDAF++ with pair2vec (single model)85.535No---
97aviqa (ensemble)85.469No---
98test85.348No---
99MEMEN (single model)85.344NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
100MEMEN (single model)85.344NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
101Interactive AoA Reader (ensemble)85.297No---
102AttentionReader+ (single)84.925No---
103DNET (single model)84.905No---
104BiDAF++ (single model)84.858No---
105MARS (single model)84.739No---
106Conductor-net (ensemble)84.63NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
107QANet + data augmentation ×384.6NoQANet: Combining Local Convolution with Global S...2018-04-23Code
108RuBERT84.6NoAdaptation of Deep Bidirectional Multilingual Tr...2019-05-17Code
109FRC (single model)84.599No---
110VS^3-NET (single model)84.491No---
111Jenga (ensemble)84.466No---
112SAN (single model)84.396NoStochastic Answer Networks for Machine Reading C...2017-12-10Code
113r-net (single model)84.265No---
114r-net (single model)84.265No---
115RaSoR + TR + LM (single model)84.163NoContextualized Word Representations for Reading ...2017-12-10Code
116Conductor-net (ensemble)83.991No---
117{gqa} (single model)83.931No---
118FusionNet (single model)83.9NoFusionNet: Fusing via Fully-Aware Attention with...2017-11-16Code
119Interactive AoA Reader+ (single model)83.843No---
120KAR (single model)83.538NoExplicit Utilization of General Knowledge in Mac...2018-09-10-
121smarnet (ensemble)83.475No---
122Kbs (single model)83.405No---
123AVIQA-v2 (single model)83.305No---
124RaSoR + TR (single model)83.261NoContextualized Word Representations for Reading ...2017-12-10Code
125EfficientQA 125M83.1NoEfficientQA : a RoBERTa Based Phrase-Indexed Que...2021-01-06-
126SLQA (single model)82.815No---
127DCN+ (single model)82.806NoDCN+: Mixed Objective and Deep Residual Coattent...2017-10-31Code
128Mixed model (ensemble)82.769No---
129Conductor-net (single model)82.742NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
130two-attention-self-attention (ensemble)82.716No---
131MEMEN (ensemble)82.658NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
132ReasoNet (ensemble)82.552YesReasoNet: Learning to Stop Reading in Machine Co...2016-09-17-
133eeAttNet (single model)82.501No---
134Mnemonic Reader (ensemble)82.371NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
135S^3-Net (ensemble)82.342No---
136Conductor-net (single)81.933NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
137Interactive AoA Reader (single model)81.931No---
138SEDT (ensemble model)81.761NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
139Jenga (single model)81.754No---
140SSAE (ensemble)81.665No---
141SEDT+BiDAF (ensemble)81.53NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
142BiDAF (ensemble)81.525NoBidirectional Attention Flow for Machine Compreh...2016-11-05Code
143jNet (ensemble)81.517NoExploring Question Understanding and Adaptation ...2017-03-14-
144Conductor-net (single)81.415No---
145Multi-Perspective Matching (ensemble)81.257NoMulti-Perspective Context Matching for Machine C...2016-12-13Code
146BiDAF + Self Attention (single model)81.048NoSimple and Effective Multi-Paragraph Reading Com...2017-10-29Code
147S^3-Net (single model)81.023No---
148two-attention-self-attention (single model)81.011No---
149T-gating (ensemble)81.001No---
150AVIQA (single model)80.55No---
151attention+self-attention (single model)80.462No---
152Dynamic Coattention Networks (ensemble)80.383NoDynamic Coattention Networks For Question Answer...2016-11-05Code
153SRU80.2NoSimple Recurrent Units for Highly Parallelizable...2017-09-08Code
154smarnet (single model)80.16NoSmarnet: Teaching Machines to Read and Comprehen...2017-10-08-
155Mnemonic Reader (single model)80.146NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
156QFASE79.989No---
157MAMCN (single model)79.939No---
158DCN + Char + CoVe79.9NoLearned in Translation: Contextualized Word Vect...2017-08-01Code
159M-NET (single)79.835No---
160jNet (single model)79.821NoExploring Question Understanding and Adaptation ...2017-03-14-
161AttReader (single)79.725No---
162Ruminating Reader (single model)79.456NoRuminating Reader: Reasoning with Gated Multi-Ho...2017-04-24-
163ReasoNet (single model)79.364NoReasoNet: Learning to Stop Reading in Machine Co...2016-09-17-
164Document Reader (single model)79.353NoReading Wikipedia to Answer Open-Domain Questions2017-03-31Code
165FastQAExt78.857NoMaking Neural QA as Simple as Possible but not S...2017-03-14Code
166Multi-Perspective Matching (single model)78.784NoMulti-Perspective Context Matching for Machine C...2016-12-13Code
167RaSoR (single model)78.741NoLearning Recurrent Span Representations for Extr...2016-11-04Code
168SSR-BiDAF78.358No---
169SimpleBaseline (single model)78.236No---
170SEDT+BiDAF (single model)77.971NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
171PQMN (single model)77.783No---
172FABIR77.605NoA Fully Attention-Based Information Retriever2018-10-22Code
173T-gating (single model)77.569No---
174SEDT (single model)77.527NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
175BiDAF (single model)77.323NoBidirectional Attention Flow for Machine Compreh...2016-11-05Code
176AllenNLP BiDAF (single model)77.151No---
177FastQA77.07NoMaking Neural QA as Simple as Possible but not S...2017-03-14Code
178Match-LSTM with Ans-Ptr (Boundary) (ensemble)77.022NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
179Iterative Co-attention Network76.786No---
180BIDAF-COMPOUND-DSS (single model)76.429No---
181BIDAF-INDEPENDENT-DSS (single model)76.349No---
182Dynamic Coattention Networks (single model)75.896NoDynamic Coattention Networks For Question Answer...2016-11-05Code
183newtest75.787No---
184BIDAF-INDEPENDENT (single model)74.594No---
185BIDAF-COMPOUND (single model)74.555No---
186Unnamed submission by ravioncodalab73.921No---
187Match-LSTM with Bi-Ans-Ptr (Boundary)73.743NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
188Attentive CNN context with LSTM73.463No---
189Fine-Grained Gating73.327NoWords or Characters? Fine-grained Gating for Rea...2016-11-06Code
190OTF dict+spelling (single)73.056NoLearning to Compute Word Embeddings On the Fly2017-06-01-
191OTF spelling (single)72.016NoLearning to Compute Word Embeddings On the Fly2017-06-01-
192OTF spelling+lemma (single)71.968NoLearning to Compute Word Embeddings On the Fly2017-06-01-
193RQA+IDR (single model)71.389NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
194RQA+IDR (single model)71.389NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
195Dynamic Chunk Reader70.956NoEnd-to-End Answer Chunk Extraction and Ranking f...2016-10-31-
196Match-LSTM with Ans-Ptr (Boundary)70.695NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
197Unnamed submission by Will_Wu69.436No---
198Match-LSTM with Ans-Ptr (Sentence)67.748NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
199RQA (single model)65.467NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
200RQA (single model)65.467NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
201UQA (single model)64.036No---
202Unnamed submission by jinhyuklee62.78No---
203Unnamed submission by minjoon62.757No---
204UnsupervisedQA V1 (ensemble)56.436No---
205UnsupervisedQA V1 (single model)54.723No---
206QANet (single model)13.211No---
2076.907No---
208QANet (ensemble)0No---
209superman-new-des0No---
210WAHnGREA0No---
211superman-des0No---
212XLNet-deep (ensemble)0No---