Machine Reading Comprehension on Molweni (test)

64.3EM

Human performance

Updated 4d ago

Evaluation Results

Method
Human performance 2021.04	64.3	80.2
Human performance 2021.10	64.3	80.2
Human performance 2022.10	64.3	80.2
QuISG 2022.10	59.32	72.86
Enhanced Speaker-aware Model 2021.09	58.6	72.2
ELECTRA_our 2021.09	58	72.9
SelfSuper 2022.10	58	72.9
Our Baselines 2021.09	57.9	71
Speaker Embedding 2021.09	57.9	71.1
MDFN 2021.09	57.9	71.1
ELECTRA (Reimpl.) 2022.10	57.85	72.17
Public Baselines 2021.09	57.3	70.4
ELECTRA_baseline 2021.09	56.8	70.6
Speaker Embedding 2021.09	56	68.3
Enhanced Speaker-aware Model 2021.09	56	69.1
MDFN 2021.09	55.8	68.7
Multi-task 2021.10	54.9	68.4
Li et al. (2020) 2021.10	54.7	67.7
Public Baselines 2021.09	54.7	67.6
Our Baselines 2021.09	53.9	67.5
QA-only 2021.10	53.8	67.5
Enhanced Speaker-aware Model 2021.09	52.9	66.9
Speaker Embedding 2021.09	52.4	65.7
Our Baselines 2021.09	52	65.6
Li et al. (2020) 2021.10	51.8	65.5
Public Baselines 2021.09	51.8	65.5
MDFN 2021.09	51.7	65.6
Multi-task 2021.10	50.6	64.9
Enhanced Speaker-aware Model 2021.09	49.7	64.4
QA-only 2021.10	49.6	64
BERT_our 2021.09	49.2	64
SelfSuperBERT 2022.10	49.2	64
MDFN 2021.09	48.4	62.4
Speaker Embedding 2021.09	47.9	61.5
Multi-task 2021.10	47.1	61.3
DADgraph 2021.04	46.5	61.5
BERT_DADGraph 2021.09	46.5	61.5
DADGraph 2022.10	46.5	61.5
QA-only 2021.10	46.2	59.2
BERT_our baseline 2021.09	45.8	60.2
DialogueGCN 2021.04	45.7	61
Our Baselines 2021.09	45.7	58.8
DialogueRNN 2021.04	45.4	60.9
BERT 2021.04	45.3	58
BERT_public baseline 2021.09	45.3	58
Li et al. (2020) 2021.10	45.3	58
Public Baselines 2021.09	45.3	58
DocQA 2021.04	42.5	56
BiDAF 2021.04	22.9	39.8