Browse › Natural Language Processing › Question Answering › SQuAD1.1
SQuAD1.1 Benchmark (Question Answering)
Syntology Since the archive (Syntology): none placed; 3 newer papers judged to report on this board so far.
Question answering can be segmented into domain-specific tasks like community question answering and knowledge-base question answering. Popular benchmark datasets for evaluation question answering systems include SQuAD, HotPotQA, bAbI, TriviaQA, WikiQA, and many others. Models for question answering are typically evaluated on metrics like EM and F1. Some recent top performing models are T5 and XLNet.
The archive carries no text for this table; the description above is the archive's text for the task Question Answering. archive 2025-07-28
Over time archive 2025-07-28
Direction inferred from the metric name, not from the archive: F1 (higher is better). Not inferred (points only, no best-so-far line): EM, Hardware Burden, Exact Match, Operations per network pass. Points are placed at the row's paper date, not an evaluation date; 76 of 213 rows carry one.
The chart needs JavaScript; the table below carries every value.
Results archive 2025-07-28
Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.
| Paper | Code | Ran Syntology | Report | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ANNA (single model) | 90.622 | 95.719 | ✓ | – | – | not matched | report | ||||
| 2 | LUKE (single model) | 90.202 | 95.379 | ✓ | – | – | not matched | report | ||||
| 3 | LUKE (single model) | 90.202 | 95.379 | – | Paper | Code | 2020 | official (archive's flag): 3 ran · 3 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified | report | |||
| 4 | LUKE | 90.2 | – | Paper | Code | 2020 | official (archive's flag): 3 ran · 3 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified | report | ||||
| 5 | XLNet (single model) | 89.898 | 95.080 | – | – | – | not matched | report | ||||
| 6 | XLNet (single model) | 89.898 | 95.080 | 46449G | ✓ | Paper | Code | 2019 | official (archive's flag): 1 ran · 15 ran (of which 2 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified | report | ||
| 7 | XLNET-123++ (single model) | 89.856 | 94.903 | – | – | – | not matched | report | ||||
| 8 | XLNET-123+ (single model) | 89.709 | 94.859 | – | – | – | not matched | report | ||||
| 9 | XLNET-123 (single model) | 89.646 | 94.930 | – | – | – | not matched | report | ||||
| 10 | Unnamed submission by NMC | 88.912 | 94.584 | – | – | – | not matched | report | ||||
| 11 | BERTSP (single model) | 88.912 | 94.584 | – | – | – | not matched | report | ||||
| 12 | SpanBERT (single model) | 88.839 | 94.635 | – | – | – | not matched | report | ||||
| 13 | SpanBERT (single model) | 88.8 | 94.6 | 586G | – | Paper | Code | 2019 | official (archive's flag): 2 ran · 9 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 6 unverified | report | ||
| 14 | BERT+WWM+MT (single model) | 88.650 | 94.393 | – | – | – | not matched | report | ||||
| 15 | Tuned BERT-1seq Large Cased (single model) | 87.465 | 93.294 | – | – | – | not matched | report | ||||
| 16 | LinkBERT (large) | 87.45 | 92.7 | – | Paper | Code | 2022 | official: harvested, nothing ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified | report | |||
| 17 | BERT (ensemble) | 87.433 | 93.160 | – | Paper | Code | 2018 | official: no sample here; runs from other or unrecorded repositories · 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified | report | |||
| 18 | BERT-LARGE (Ensemble+TriviaQA) | 87.4 | 93.2 | – | Paper | Code | 2018 | official: no sample here; runs from other or unrecorded repositories · 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified | report | |||
| 19 | ATB (single model) | 86.940 | 92.641 | – | – | – | not matched | report | ||||
| 20 | Tuned BERT Large Cased (single model) | 86.521 | 92.617 | – | – | – | not matched | report | ||||
| 21 | BERT+MT (single model) | 86.458 | 92.645 | – | – | – | not matched | report | ||||
| 22 | Knowledge-enhanced BERT (single model) | 85.944 | 92.425 | – | – | – | not matched | report | ||||
| 23 | KT-NET (single model) | 85.944 | 92.425 | – | – | – | not matched | report | ||||
| 24 | ST_bl | 85.430 | 91.976 | – | – | – | not matched | report | ||||
| 25 | nlnet (ensemble) | 85.356 | 91.202 | – | – | – | not matched | report | ||||
| 26 | EL-BERT (single model) | 85.335 | 91.807 | – | – | – | not matched | report | ||||
| 27 | BISAN (single model) | 85.314 | 91.756 | – | – | – | not matched | report | ||||
| 28 | BERT+Sparse-Transformer | 85.125 | 91.623 | – | – | – | not matched | report | ||||
| 29 | BERT (single model) | 85.083 | 91.835 | ✓ | Paper | Code | 2018 | official: no sample here; runs from other or unrecorded repositories · 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified | report | |||
| 30 | DPN (single model) | 84.978 | 92.019 | – | – | – | not matched | report | ||||
| 31 | BERT-uncased (single model) | 84.926 | 91.932 | – | – | – | not matched | report | ||||
| 32 | WD (single model) | 84.402 | 90.561 | – | – | – | not matched | report | ||||
| 33 | Original BERT Large Cased (single model) | 84.328 | 91.281 | – | – | – | not matched | report | ||||
| 34 | MARS (ensemble) | 83.982 | 89.796 | – | – | – | not matched | report | ||||
| 35 | Common-sense Governed BERT-123 (single model) | 83.930 | 90.613 | – | – | – | not matched | report | ||||
| 36 | WD1 (single model) | 83.804 | 90.429 | – | – | – | not matched | report | ||||
| 37 | nlnet (single model) | 83.468 | 90.133 | – | – | – | not matched | report | ||||
| 38 | Pytalk + Stanza + BERT (single model) | 83.426 | 89.218 | – | – | – | not matched | report | ||||
| 39 | Reinforced Mnemonic Reader + A2D (ensemble model) | 82.849 | 88.764 | – | – | – | not matched | report | ||||
| 40 | BERT-Base mod (single model) | 82.681 | 89.379 | – | – | – | not matched | report | ||||
| 41 | r-net+ (ensemble) | 82.650 | 88.493 | – | – | – | not matched | report | ||||
| 42 | Hybrid AoA Reader (ensemble) | 82.482 | 89.281 | – | – | – | not matched | report | ||||
| 43 | QANet (single) | 82.471 | 89.306 | – | – | – | not matched | report | ||||
| 44 | SLQA+ (ensemble) | 82.440 | 88.607 | – | – | – | not matched | report | ||||
| 45 | Reinforced Mnemonic Reader (ensemble model) | 82.283 | 88.533 | ✓ | Paper | Code | 2017 | linked, not harvested | report | |||
| 46 | r-net (ensemble) | 82.136 | 88.126 | – | – | – | not matched | report | ||||
| 47 | BERT (single model) | 82.062 | 88.947 | – | – | – | not matched | report | ||||
| 48 | AttentionReader+ (ensemble) | 81.790 | 88.163 | – | – | – | not matched | report | ||||
| 49 | MMIPN | 81.580 | 88.948 | – | – | – | not matched | report | ||||
| 50 | BERT - 6 Layers | 81.5 | 88.5 | – | Paper | Code | 2021 | linked, not harvested | report | |||
| 51 | KACTEIL-MRC(GF-Net+) (ensemble) | 81.496 | 87.557 | – | – | – | not matched | report | ||||
| 52 | Reinforced Mnemonic Reader + A2D + DA (single model) | 81.401 | 88.122 | – | – | – | not matched | report | ||||
| 53 | ARSG-BERT (single model) | 81.307 | 88.909 | – | – | – | not matched | report | ||||
| 54 | BERT-COMPOUND-DSS (single model) | 81.045 | 87.999 | – | – | – | not matched | report | ||||
| 55 | BiDAF + Self Attention + ELMo (ensemble) | 81.003 | 87.432 | – | Paper | Code | 2018 | 30 ran (of which 12 constructed an object rather than computing a result; 24 with no instrument failure: 2 honoured, 0 violated, 22 with no contract checked; 6 where Syntology's instrument failed) · 28 unverified | report | |||
| 56 | BiDAF + Self Attention + ELMo (ensemble) | 81.003 | 87.432 | – | – | – | not matched | report | ||||
| 57 | BERT-COMPOUND (single model) | 80.720 | 87.758 | – | – | – | not matched | report | ||||
| 58 | mBERT + Task Adapter (Single) | 80.667 | 88.169 | – | – | – | not matched | report | ||||
| 59 | AVIQA+ (ensemble) | 80.615 | 87.311 | – | – | – | not matched | report | ||||
| 60 | Reinforced Mnemonic Reader + A2D (single model) | 80.489 | 87.454 | – | – | – | not matched | report | ||||
| 61 | SLQA+ | 80.436 | 87.021 | – | – | – | not matched | report | ||||
| 62 | EAZI (ensemble) | 80.436 | 86.912 | – | – | – | not matched | report | ||||
| 63 | EAZI+ (ensemble) | 80.426 | 86.912 | – | – | – | not matched | report | ||||
| 64 | DNET (ensemble) | 80.164 | 86.721 | – | – | – | not matched | report | ||||
| 65 | Hybrid AoA Reader (single model) | 80.027 | 87.288 | – | – | – | not matched | report | ||||
| 66 | BiDAF + Self Attention + ELMo + A2D (single model) | 79.996 | 86.711 | – | – | – | not matched | report | ||||
| 67 | r-net+ (single model) | 79.901 | 86.536 | – | – | – | not matched | report | ||||
| 68 | batch (single model) | 79.859 | 88.263 | – | – | – | not matched | report | ||||
| 69 | MAMCN+ (single model) | 79.692 | 86.727 | – | Paper | – | 2018 | no code linked | report | |||
| 70 | MAMCN+ (single model) | 79.692 | 86.727 | – | – | – | not matched | report | ||||
| 71 | SAN (ensemble model) | 79.608 | 86.496 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 72 | BERT-INDEPENDENT-DSS-FILTERED (single model) | 79.597 | 87.374 | – | – | – | not matched | report | ||||
| 73 | Reinforced Mnemonic Reader (single model) | 79.545 | 86.654 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 74 | SLQA+ (single model) | 79.199 | 86.590 | – | – | – | not matched | report | ||||
| 75 | Interactive AoA Reader+ (ensemble) | 79.083 | 86.450 | – | – | – | not matched | report | ||||
| 76 | MIR-MRC(F-Net) (single model) | 79.083 | 86.288 | – | – | – | not matched | report | ||||
| 77 | KACTEIL-MRC(GF-Net+Distillation) (single model) | 79.083 | 86.288 | – | – | – | not matched | report | ||||
| 78 | KACTEIL-MRC (GF-Net+Distillation) | 79.083 | 86.288 | – | – | – | not matched | report | ||||
| 79 | MDReader | 79.031 | 86.006 | – | – | – | not matched | report | ||||
| 80 | FusionNet (ensemble) | 78.978 | 86.016 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 81 | DCN+ (ensemble) | 78.852 | 85.996 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 82 | KACTEIL-MRC(GF-Net+) (single model) | 78.664 | 85.780 | – | – | – | not matched | report | ||||
| 83 | KACTEIL-MRC (GF-Net+) | 78.664 | 85.780 | – | – | – | not matched | report | ||||
| 84 | BERT-INDEPENDENT (single model) | 78.653 | 86.663 | – | – | – | not matched | report | ||||
| 85 | BiDAF + Self Attention + ELMo (single model) | 78.58 | 85.833 | – | Paper | Code | 2018 | 30 ran (of which 12 constructed an object rather than computing a result; 24 with no instrument failure: 2 honoured, 0 violated, 22 with no contract checked; 6 where Syntology's instrument failed) · 28 unverified | report | |||
| 86 | BiDAF + Self Attention + ELMo (single model) | 78.580 | 85.833 | – | – | – | not matched | report | ||||
| 87 | aviqa (ensemble) | 78.496 | 85.469 | – | – | – | not matched | report | ||||
| 88 | KakaoNet (single model) | 78.401 | 85.724 | – | – | – | not matched | report | ||||
| 89 | SLQA(ensemble) | 78.328 | 85.682 | – | – | – | not matched | report | ||||
| 90 | SLQA (ensemble) | 78.328 | 85.682 | – | – | – | not matched | report | ||||
| 91 | MEMEN (single model) | 78.234 | 85.344 | – | Paper | – | 2017 | no code linked | report | |||
| 92 | MEMEN (single model) | 78.234 | 85.344 | – | Paper | – | 2017 | no code linked | report | |||
| 93 | BiDAF++ with pair2vec (single model) | 78.223 | 85.535 | – | – | – | not matched | report | ||||
| 94 | MDReader0 | 78.171 | 85.543 | – | – | – | not matched | report | ||||
| 95 | test | 78.087 | 85.348 | – | – | – | not matched | report | ||||
| 96 | Interactive AoA Reader (ensemble) | 77.845 | 85.297 | – | – | – | not matched | report | ||||
| 97 | BERT - 3 Layers | 77.7 | 85.8 | – | Paper | Code | 2021 | linked, not harvested | report | |||
| 98 | DNET (single model) | 77.646 | 84.905 | – | – | – | not matched | report | ||||
| 99 | RaSoR + TR + LM (single model) | 77.583 | 84.163 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 100 | BiDAF++ (single model) | 77.573 | 84.858 | – | – | – | not matched | report | ||||
| 101 | AttentionReader+ (single) | 77.342 | 84.925 | – | – | – | not matched | report | ||||
| 102 | Jenga (ensemble) | 77.237 | 84.466 | – | – | – | not matched | report | ||||
| 103 | {gqa} (single model) | 77.090 | 83.931 | – | – | – | not matched | report | ||||
| 104 | Conductor-net (ensemble) | 76.996 | 84.630 | – | Paper | – | 2017 | no code linked | report | |||
| 105 | MARS (single model) | 76.859 | 84.739 | – | – | – | not matched | report | ||||
| 106 | SAN (single model) | 76.828 | 84.396 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 107 | VS^3-NET (single model) | 76.775 | 84.491 | – | – | – | not matched | report | ||||
| 108 | r-net (single model) | 76.461 | 84.265 | – | – | – | not matched | report | ||||
| 109 | r-net (single model) | 76.461 | 84.265 | – | Paper | – | 2017 | no code linked | report | |||
| 110 | FRC (single model) | 76.240 | 84.599 | – | – | – | not matched | report | ||||
| 111 | QANet + data augmentation ×3 | 76.2 | 84.6 | – | Paper | Code | 2018 | 14 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 3 honoured, 0 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified | report | |||
| 112 | Conductor-net (ensemble) | 76.146 | 83.991 | – | – | – | not matched | report | ||||
| 113 | KAR (single model) | 76.125 | 83.538 | – | Paper | – | 2018 | no code linked | report | |||
| 114 | smarnet (ensemble) | 75.989 | 83.475 | – | – | – | not matched | report | ||||
| 115 | FusionNet (single model) | 75.968 | 83.900 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 116 | AVIQA-v2 (single model) | 75.926 | 83.305 | – | – | – | not matched | report | ||||
| 117 | Interactive AoA Reader+ (single model) | 75.821 | 83.843 | – | – | – | not matched | report | ||||
| 118 | RaSoR + TR (single model) | 75.789 | 83.261 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 119 | MEMEN (ensemble) | 75.370 | 82.658 | – | Paper | – | 2017 | no code linked | report | |||
| 120 | Mixed model (ensemble) | 75.265 | 82.769 | – | – | – | not matched | report | ||||
| 121 | two-attention-self-attention (ensemble) | 75.223 | 82.716 | – | – | – | not matched | report | ||||
| 122 | Kbs (single model) | 75.034 | 83.405 | – | – | – | not matched | report | ||||
| 123 | ReasoNet (ensemble) | 75.034 | 82.552 | ✓ | Paper | – | 2016 | no code linked | report | |||
| 124 | EfficientQA 125M | 74.9 | 83.1 | – | Paper | – | 2021 | no code linked | report | |||
| 125 | DCN+ (single model) | 74.866 | 82.806 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 126 | eeAttNet (single model) | 74.604 | 82.501 | – | – | – | not matched | report | ||||
| 127 | SLQA (single model) | 74.489 | 82.815 | – | – | – | not matched | report | ||||
| 128 | Conductor-net (single model) | 74.405 | 82.742 | – | Paper | – | 2017 | no code linked | report | |||
| 129 | Mnemonic Reader (ensemble) | 74.268 | 82.371 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 130 | S^3-Net (ensemble) | 74.121 | 82.342 | – | – | – | not matched | report | ||||
| 131 | SEDT (ensemble model) | 74.090 | 81.761 | – | Paper | – | 2017 | no code linked | report | |||
| 132 | SSAE (ensemble) | 74.080 | 81.665 | – | – | – | not matched | report | ||||
| 133 | Multi-Perspective Matching (ensemble) | 73.765 | 81.257 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 134 | BiDAF (ensemble) | 73.744 | 81.525 | – | Paper | Code | 2016 | official (archive's flag): 1 ran · 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 2 violated, 1 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified | report | |||
| 135 | SEDT+BiDAF (ensemble) | 73.723 | 81.530 | – | Paper | – | 2017 | no code linked | report | |||
| 136 | Interactive AoA Reader (single model) | 73.639 | 81.931 | – | – | – | not matched | report | ||||
| 137 | Jenga (single model) | 73.303 | 81.754 | – | – | – | not matched | report | ||||
| 138 | Conductor-net (single) | 73.240 | 81.933 | – | Paper | – | 2017 | no code linked | report | |||
| 139 | jNet (ensemble) | 73.010 | 81.517 | – | Paper | – | 2017 | no code linked | report | |||
| 140 | T-gating (ensemble) | 72.758 | 81.001 | – | – | – | not matched | report | ||||
| 141 | two-attention-self-attention (single model) | 72.600 | 81.011 | – | – | – | not matched | report | ||||
| 142 | Conductor-net (single) | 72.590 | 81.415 | – | – | – | not matched | report | ||||
| 143 | AVIQA (single model) | 72.485 | 80.550 | – | – | – | not matched | report | ||||
| 144 | BiDAF + Self Attention (single model) | 72.139 | 81.048 | – | Paper | Code | 2017 | official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified | report | |||
| 145 | S^3-Net (single model) | 71.908 | 81.023 | – | – | – | not matched | report | ||||
| 146 | QFASE | 71.898 | 79.989 | – | – | – | not matched | report | ||||
| 147 | attention+self-attention (single model) | 71.698 | 80.462 | – | – | – | not matched | report | ||||
| 148 | Dynamic Coattention Networks (ensemble) | 71.625 | 80.383 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 149 | smarnet (single model) | 71.415 | 80.160 | – | Paper | – | 2017 | no code linked | report | |||
| 150 | SRU | 71.4 | 80.2 | 4G | – | Paper | Code | 2017 | community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified | report | ||
| 151 | AttReader (single) | 71.373 | 79.725 | – | – | – | not matched | report | ||||
| 152 | DCN + Char + CoVe | 71.3 | 79.9 | – | Paper | Code | 2017 | community repositories only · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified | report | |||
| 153 | M-NET (single) | 71.016 | 79.835 | – | – | – | not matched | report | ||||
| 154 | Mnemonic Reader (single model) | 70.995 | 80.146 | – | Paper | Code | 2017 | linked, not harvested | report | |||
| 155 | MAMCN (single model) | 70.985 | 79.939 | – | – | – | not matched | report | ||||
| 156 | FastQAExt | 70.849 | 78.857 | – | Paper | Code | 2017 | 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified | report | |||
| 157 | RaSoR (single model) | 70.849 | 78.741 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 158 | Document Reader (single model) | 70.733 | 79.353 | – | Paper | Code | 2017 | community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified | report | |||
| 159 | Ruminating Reader (single model) | 70.639 | 79.456 | – | Paper | – | 2017 | no code linked | report | |||
| 160 | jNet (single model) | 70.607 | 79.821 | – | Paper | – | 2017 | no code linked | report | |||
| 161 | ReasoNet (single model) | 70.555 | 79.364 | – | Paper | – | 2016 | no code linked | report | |||
| 162 | Multi-Perspective Matching (single model) | 70.387 | 78.784 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 163 | SimpleBaseline (single model) | 69.600 | 78.236 | – | – | – | not matched | report | ||||
| 164 | SSR-BiDAF | 69.443 | 78.358 | – | – | – | not matched | report | ||||
| 165 | SEDT+BiDAF (single model) | 68.478 | 77.971 | – | Paper | – | 2017 | no code linked | report | |||
| 166 | FastQA | 68.436 | 77.070 | – | Paper | Code | 2017 | 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified | report | |||
| 167 | PQMN (single model) | 68.331 | 77.783 | – | – | – | not matched | report | ||||
| 168 | SEDT (single model) | 68.163 | 77.527 | – | Paper | – | 2017 | no code linked | report | |||
| 169 | T-gating (single model) | 68.132 | 77.569 | – | – | – | not matched | report | ||||
| 170 | BiDAF (single model) | 67.974 | 77.323 | – | Paper | Code | 2016 | official (archive's flag): 1 ran · 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 2 violated, 1 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified | report | |||
| 171 | Match-LSTM with Ans-Ptr (Boundary) (ensemble) | 67.901 | 77.022 | – | Paper | Code | 2016 | official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified | report | |||
| 172 | FABIR | 67.744 | 77.605 | – | Paper | Code | 2018 | linked, not harvested | report | |||
| 173 | AllenNLP BiDAF (single model) | 67.618 | 77.151 | – | – | – | not matched | report | ||||
| 174 | BIDAF-COMPOUND-DSS (single model) | 67.544 | 76.429 | – | – | – | not matched | report | ||||
| 175 | Iterative Co-attention Network | 67.502 | 76.786 | – | – | – | not matched | report | ||||
| 176 | newtest | 66.527 | 75.787 | – | – | – | not matched | report | ||||
| 177 | BIDAF-INDEPENDENT-DSS (single model) | 66.516 | 76.349 | – | – | – | not matched | report | ||||
| 178 | Dynamic Coattention Networks (single model) | 66.233 | 75.896 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 179 | BIDAF-COMPOUND (single model) | 65.163 | 74.555 | – | – | – | not matched | report | ||||
| 180 | BIDAF-INDEPENDENT (single model) | 64.932 | 74.594 | – | – | – | not matched | report | ||||
| 181 | Match-LSTM with Bi-Ans-Ptr (Boundary) | 64.744 | 73.743 | – | Paper | Code | 2016 | official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified | report | |||
| 182 | Unnamed submission by ravioncodalab | 64.439 | 73.921 | – | – | – | not matched | report | ||||
| 183 | OTF dict+spelling (single) | 64.083 | 73.056 | – | Paper | – | 2017 | no code linked | report | |||
| 184 | Attentive CNN context with LSTM | 63.306 | 73.463 | – | – | – | not matched | report | ||||
| 185 | OTF spelling (single) | 62.897 | 72.016 | – | Paper | – | 2017 | no code linked | report | |||
| 186 | OTF spelling+lemma (single) | 62.604 | 71.968 | – | Paper | – | 2017 | no code linked | report | |||
| 187 | Dynamic Chunk Reader | 62.499 | 70.956 | – | Paper | – | 2016 | no code linked | report | |||
| 188 | Fine-Grained Gating | 62.446 | 73.327 | – | Paper | Code | 2016 | linked, not harvested | report | |||
| 189 | RQA+IDR (single model) | 61.145 | 71.389 | – | – | – | not matched | report | ||||
| 190 | RQA+IDR (single model) | 61.145 | 71.389 | – | Paper | Code | 2020 | linked, not harvested | report | |||
| 191 | Match-LSTM with Ans-Ptr (Boundary) | 60.474 | 70.695 | – | Paper | Code | 2016 | official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified | report | |||
| 192 | Unnamed submission by Will_Wu | 59.058 | 69.436 | – | – | – | not matched | report | ||||
| 193 | RQA (single model) | 55.827 | 65.467 | – | – | – | not matched | report | ||||
| 194 | RQA (single model) | 55.827 | 65.467 | – | Paper | Code | 2020 | linked, not harvested | report | |||
| 195 | Match-LSTM with Ans-Ptr (Sentence) | 54.505 | 67.748 | – | Paper | Code | 2016 | official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified | report | |||
| 196 | UQA (single model) | 53.698 | 64.036 | – | – | – | not matched | report | ||||
| 197 | Unnamed submission by jinhyuklee | 52.544 | 62.780 | – | – | – | not matched | report | ||||
| 198 | Unnamed submission by minjoon | 52.533 | 62.757 | – | – | – | not matched | report | ||||
| 199 | UnsupervisedQA V1 (ensemble) | 47.341 | 56.436 | – | – | – | not matched | report | ||||
| 200 | UnsupervisedQA V1 (single model) | 44.215 | 54.723 | – | – | – | not matched | report | ||||
| 201 | QANet (single model) | 12.273 | 13.211 | – | – | – | not matched | report | ||||
| 202 | (unnamed in the archive) | 0.000 | 6.907 | – | – | – | not matched | report | ||||
| 203 | QANet (ensemble) | 0.000 | 0.000 | – | – | – | not matched | report | ||||
| 204 | superman-new-des | 0.000 | 0.000 | – | – | – | not matched | report | ||||
| 205 | WAHnGREA | 0.000 | 0.000 | – | – | – | not matched | report | ||||
| 206 | superman-des | 0.000 | 0.000 | – | – | – | not matched | report | ||||
| 207 | XLNet-deep (ensemble) | 0.000 | 0.000 | – | – | – | not matched | report | ||||
| 208 | LUKE 483M | 95.4 | – | Paper | Code | 2020 | official (archive's flag): 3 ran · 3 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified | report | ||||
| 209 | BART (TextBox 2.0) | 93.04 | 86.44 | – | Paper | Code | 2022 | official (archive's flag): 1 ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result | report | |||
| 210 | BERT-LARGE (Single+TriviaQA) | 91.8 | – | Paper | Code | 2018 | official: no sample here; runs from other or unrecorded repositories · 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified | report | ||||
| 211 | BERT-Large 32k batch size with AdamW | 91.58 | – | Paper | – | 2021 | no code linked | report | ||||
| 212 | DyREX | 91.01 | – | Paper | Code | 2022 | official (archive's flag): 1 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified | report | ||||
| 213 | RuBERT | 84.6 | – | Paper | Code | 2019 | linked, not harvested | report |
All 213 rows shown. 76 link to a paper page on this site; 6 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28
Syntology Ran reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”: of the code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and U are unverified (harvested, no recorded run). Of the N, C constructed an object rather than computing a result (a part of N, never taken away from it), K ran with no instrument failure (honoured or violated the contract Syntology drafted, or ran with no contract checked) and I are failures of Syntology's instrument, not of the code (a wrong contract draft, or a contract check that could not run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 27 rows have a graph line, from 16 distinct papers. Rows whose paper has a sample that ran: 27 (27 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument). Papers: 16 with a sample that ran: 16 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument. Whose code ran. When the archive marks a repository official for the row's paper, the cell starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code): “official (archive's flag): N ran” counts the samples from it that ran, instrument failures included; “official: harvested, nothing ran”; when it has no sample for this paper and other samples ran, “community repositories only” if every one came from a repository the archive lists for the paper but neither marks official nor records as mentioned in it, or one Syntology harvested outside the archive's links, and “official: no sample here; runs from other or unrecorded repositories” if any came from a repository named in the paper or found in its text by Syntology, or has no repository recorded; “official: harvested for another paper” (no sample from it is linked to this paper) or “official: not harvested”. No state is printed when the archive marks none official. The cell's tooltip names the repositories the samples that ran came from. Of the 27 rows with a graph line, 22 have an archive-official repository: on 10 a sample from it ran (9 with a run with no instrument failure), on 3 the samples that ran came from community repositories only, and on 8 from other or unrecorded repositories. The filters above hide rows and never renumber them; the first two count, by default, only a run with no instrument failure, and offer any run, instrument failures included. Counting each paper once, of 828 samples: 397 ran (of which 92 constructed an object rather than computing a result; 304 with no instrument failure: 30 honoured, 9 violated, 265 with no contract checked; 93 where Syntology's instrument failed) · 431 unverified. Separately, 218 of those 828 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. Per-sample status is on the paper page; each Ran cell links to that paper's sample list. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.
Since the archive Syntology
Of the 8,888 newer papers whose tables Syntology has read (8,886 checked in full and 2 with a leaderboard still awaiting a verdict), 27 had this board among the leaderboards the extractor considered for their tables (it keeps at most 10 boards per paper and 4 per dataset; the lists of 1,575 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 3 proposed a result; of those 3, 2 gave no cell that passed the extractor's own checks, 1 was rejected by the independent check, 0 await a verdict and 0 were placed. Syntology has checked 8,886 of the 18,637 papers on this site that are newer than the archive (for 3,717 of them no archive leaderboard matched the paper's tables, so there was nothing further to check); 733 were read and have nothing to place (arXiv has no HTML version of the paper, or that version has no tables), 41 could not be read (the extractor's reply could not be parsed), and results from the other 8,977 appear after they are checked.
Newer papers whose tables Claude Sonnet 4.5 judged to report on this benchmark (a model's judgement, not a result; newest first by arXiv id)
Syntology 3 newer papers judged to report on this board: the papers counted above as having proposed a result (0 placed); every one is listed here and in the JSON twin. The order is the arXiv identifier's, newest first, not a rank. The judgement is the model's, read from the paper's own tables; the outcome is what Syntology's checks did with it, and no value from a paper that was not placed is printed. “refused by a rule” names the extractor's rule; “no cell passed the extractor's checks” means the extractor's record holds no cell for this board, passed or refused. The 2 counted above as giving no cell that passed the extractor's own checks are the 2 marked “refused by a rule”. Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections