Question Answering On Squad20

Metriken

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Vergleichstabelle

Modellname	EM	F1
Modell 1	80.208	83.149
Modell 2	79.971	83.184
Modell 3	80.038	82.796
Modell 4	80.117	83.189
Modell 5	82.803	85.863
Modell 6	74.769	77.706
Modell 7	89.021	91.765
albert-a-lite-bert-for-self-supervised	88.107	90.902
Modell 9	80.377	83.262
Modell 10	86.436	89.086
Modell 11	75.344	78.381
Modell 12	84.800	87.864
Modell 13	73.742	76.858
Modell 14	80.388	82.908
Modell 15	78.933	81.863
Modell 16	78.052	81.174
Modell 17	90.002	92.497
Modell 18	84.924	88.204
ensemble-albert-on-squad-2-0	-	90.123
Modell 20	79.181	82.259
Modell 21	90.487	92.894
Modell 22	85.838	88.921
Modell 23	82.995	86.035
Modell 24	84.642	88.000
Modell 25	89.551	92.366
Modell 26	77.003	80.209
Modell 27	89.562	92.226
Modell 28	87.847	90.532
Modell 29	86.448	89.586
Modell 30	87.802	90.872
deep-contextualized-word-representations	63.372	66.251
Modell 32	84.642	88.000
Modell 33	90.600	92.899
Modell 34	79.948	83.023
Modell 35	78.357	81.500
Modell 36	90.679	92.948
Modell 37	90.194	92.594
Modell 38	80.411	83.457
Modell 39	88.107	90.902
Modell 40	82.126	84.820
Modell 41	88.614	91.303
Modell 42	87.429	90.163
Modell 43	90.284	92.691
Modell 44	88.050	90.645
Modell 45	82.024	84.854
read-verify-machine-reading-comprehension	71.767	74.295
Modell 47	89.461	92.134
Modell 48	87.949	90.818
sg-net-syntax-guided-machine-reading	88.174	90.702
deberta-decoding-enhanced-bert-with	88.0	90.7
Modell 51	58.508	62.045
Modell 52	59.332	62.305
Modell 53	82.713	85.584
Modell 54	88.637	91.230
Modell 55	79.779	82.912
Modell 56	76.055	79.329
Modell 57	90.420	92.799
Modell 58	83.142	85.873
Modell 59	81.979	84.846
Modell 60	72.884	76.217
Modell 61	85.003	87.833
Modell 62	88.998	91.635
Modell 63	81.178	84.251
Modell 64	78.594	81.445
Modell 65	63.372	66.251
xlnet-generalized-autoregressive-pretraining	87.926	90.689
Modell 67	69.262	72.642
Modell 68	65.651	68.866
Modell 69	71.666	75.457
Modell 70	88.197	90.830
Modell 71	84.721	87.117
Modell 72	90.115	92.580
Modell 73	72.884	76.217
stochastic-answer-networks-for-machine	68.653	71.439
Modell 75	76.055	79.329
semantics-aware-bert-for-language	84.800	87.864
Modell 77	87.847	91.265
Modell 78	85.229	87.926
Modell 79	88.186	90.939
Modell 80	88.050	91.036
Modell 81	86.346	89.133
Modell 82	85.240	87.901
Modell 83	79.632	82.852
Modell 84	80.715	83.827
Modell 85	88.592	90.859
Modell 86	80.456	83.509
semantics-aware-bert-for-language	86.166	88.886
Modell 88	84.123	87.013
Modell 89	79.971	83.266
Modell 90	69.476	72.857
Modell 91	77.262	80.258
Modell 92	84.620	87.625
Modell 93	84.721	87.117
Modell 94	86.820	89.795
Modell 95	89.325	91.994
Modell 96	80.343	83.243
Modell 97	84.202	86.767
Modell 98	0.068	3.971
Modell 99	84.292	86.967
Modell 100	78.481	81.531
Modell 101	86.651	89.595
Modell 102	85.173	88.425
Modell 103	80.354	83.329
Modell 104	86.166	88.886
Modell 105	89.348	91.985
Modell 106	89.224	91.853
Modell 107	68.213	70.878
semantics-aware-bert-for-language	86.166	88.886
Modell 109	83.819	86.669
Modell 110	84.123	87.013
Modell 111	40.397	43.213
Modell 112	84.834	87.644
Modell 113	80.140	82.962
Modell 114	56.545	59.546
Modell 115	88.716	91.365
Modell 116	90.442	92.877
Modell 117	83.457	86.122
Modell 118	85.884	88.621
Modell 119	89.449	92.118
Modell 120	48.883	48.883
Modell 121	75.073	77.805
Modell 122	90.724	93.011
Modell 123	85.872	88.793
Modell 124	88.524	91.256
Modell 125	87.147	89.474
Modell 126	78.357	81.500
Modell 127	67.897	70.884
Modell 128	63.327	66.633
Modell 129	88.107	90.902
Modell 130	83.751	86.594
Modell 131	90.386	92.777
sg-net-syntax-guided-machine-reading	87.238	90.071
Modell 133	83.051	85.737
Modell 134	74.272	77.052
Modell 135	90.871	93.183
Modell 136	76.563	79.776
Modell 137	88.231	90.713
Modell 138	78.650	81.497
Modell 139	68.213	70.878
Modell 140	86.730	89.286
Modell 141	44.945	47.994
Modell 142	75.457	78.232
Modell 143	83.536	86.096
Modell 144	89.449	92.118
Modell 145	90.454	92.748
Modell 146	84.642	88.000
Modell 147	85.827	89.778
Modell 148	80.241	83.175
Modell 149	74.791	77.988
Modell 150	4.830	5.920
Modell 151	82.374	85.310
u-net-machine-reading-comprehension-with	71.417	74.869
Modell 153	80.354	83.329
Modell 154	78.876	82.524
Modell 155	74.329	77.396
Modell 156	85.748	88.709
pay-attention-to-mlps	-	78.3
Modell 158	81.731	84.862
spanbert-improving-pre-training-by	85.7	88.7
Modell 160	88.298	91.078
Modell 161	88.761	91.745
Modell 162	79.745	83.020
Modell 163	48.804	48.815
Modell 164	76.710	79.659
Modell 165	74.656	77.404
luke-deep-contextualized-entity	-	90.2
Modell 167	85.150	87.715
stochastic-answer-networks-for-machine	71.316	73.704
Modell 169	79.779	83.099
Modell 170	85.827	88.699
Modell 171	90.939	93.214
Modell 172	78.052	81.174
Modell 173	88.569	91.287
Modell 174	74.577	77.464
Modell 175	59.174	62.093
Modell 176	88.434	90.918
Modell 177	63.338	67.422
Modell 178	84.123	87.013
Modell 179	87.700	90.588
Modell 180	73.099	76.236
Modell 181	87.994	90.944
Modell 182	88.107	90.902
Modell 183	86.933	90.037
fusionnet-fusing-via-fully-aware-attention	70.300	72.484
Modell 185	86.211	88.848
Modell 186	87.046	89.899
Modell 187	78.650	81.474
Modell 188	89.743	92.180
Modell 189	68.766	71.662
Modell 190	90.081	92.457
Modell 191	82.431	85.178
Modell 192	78.933	81.863
Modell 193	89.923	92.425
sg-net-syntax-guided-machine-reading	86.211	88.848
Modell 195	87.193	89.934
Modell 196	88.851	91.486
Modell 197	88.603	91.299
Modell 198	89.528	92.059
Modell 199	84.202	86.767
Modell 200	86.673	89.147
Modell 201	83.040	85.892
Modell 202	80.896	83.604
Modell 203	88.107	91.419
Modell 204	72.670	75.507
Modell 205	86.594	89.082
Modell 206	90.521	92.824
Modell 207	78.933	81.863
Modell 208	83.142	85.873
Modell 209	79.993	83.039
Modell 210	82.882	86.002
Modell 211	88.073	91.179
Modell 212	77.262	80.258
Modell 213	82.577	85.603
retrospective-reader-for-machine-reading	90.578	92.978
roberta-a-robustly-optimized-bert-pretraining	86.820	89.795
Modell 216	88.874	91.546
Modell 217	49.695	49.701
Modell 218	87.994	90.944
Modell 219	90.059	92.517
Modell 220	84.823	87.489
Modell 221	89.133	91.666
Modell 222	90.420	92.807
Modell 223	77.262	80.258
Modell 224	82.126	84.624
Modell 225	88.592	91.286
Modell 226	90.126	92.535
Modell 227	85.703	88.400
Modell 228	57.707	62.341
Modell 229	83.469	86.043
albert-a-lite-bert-for-self-supervised	89.731	92.215
Modell 231	83.525	86.222
Modell 232	83.119	85.510
Modell 233	86.572	89.063
Modell 234	86.572	89.063
Modell 235	74.385	77.308
Modell 236	90.002	92.425
Modell 237	69.476	72.857
Modell 238	27.217	29.597
Modell 239	72.072	75.513
Modell 240	74.791	77.988
Modell 241	90.860	93.100
Modell 242	88.355	91.019
Modell 243	85.850	88.449
Modell 244	86.098	89.634
luke-deep-contextualized-entity	87.429	90.163
Modell 246	82.724	85.491
Modell 247	70.763	74.449
Modell 248	86.166	88.886
Modell 249	80.749	83.851
Modell 250	65.256	69.206
Modell 251	81.110	84.386
Modell 252	71.462	74.434
Modell 253	82.972	85.810
Modell 254	66.610	70.303
sg-net-syntax-guided-machine-reading	85.229	87.926
Modell 256	68.021	71.583
Modell 257	89.235	91.900
Modell 258	87.870	90.823
Modell 259	73.505	76.424
Modell 260	85.838	88.921
retrospective-reader-for-machine-reading	89.562	92.052
Modell 262	89.404	91.964
Modell 263	82.306	85.670
Modell 264	74.746	78.227
retrospective-reader-for-machine-reading	90.115	92.580
Modell 266	89.235	91.739
Modell 267	85.838	88.921
Modell 268	56.545	59.546
Modell 269	71.699	74.430
Modell 270	72.072	75.513
Modell 271	85.082	87.615
Modell 272	81.573	84.535
Modell 273	77.319	80.310
Modell 274	89.325	91.939
Modell 275	71.293	74.578
Modell 276	80.591	83.391
Modell 277	82.961	86.075
Modell 278	80.422	83.118
Modell 279	88.998	91.635
Modell 280	86.403	89.148
Modell 281	85.872	88.989
Modell 282	65.719	69.381
Modell 283	89.777	92.312
Modell 284	80.005	83.208
Modell 285	73.302	76.284
retrospective-reader-for-machine-reading	88.107	91.419