paper-with-me

Question Answering 벤치마크

Question Answering on SQuAD2.0 dev

13개 결과 · ⬇ CSV · JSON

F1

73.4 77.7 82 86.3 90.6 2018-08 2026-09 RMR + ELMo (Model-III) — 74.8 (2018-08-17) U-Net — 74.0 (2018-10-12) XLNet (single model) — 90.6 (2019-06-19) SpanBERT — 86.8 (2019-07-24) RoBERTa (no data aug) — 89.4 (2019-07-26) SG-Net — 87.9 (2019-08-14) SemBERT large — 83.6 (2019-09-05) TinyBERT-6 67M — 73.4 (2019-09-23) ALBERT xxlarge — 88.1 (2019-09-26) ALBERT xlarge — 85.9 (2019-09-26) ALBERT large — 82.1 (2019-09-26) ALBERT base — 79.1 (2019-09-26) XLNet+DSC — 89.51 (2019-11-07) RMR + ELMo (Model-III) — 74.8 (2018-08-17) XLNet (single model) — 90.6 (2019-06-19)
RankModel F1EM Extra Training Data PaperCodeYear
1 XLNet (single model) 90.687.9 XLNet: Generalized Autoregressive Pretraining for Language Understanding huggingface/transformers · PaddlePaddle/PaddleNLP · zihangdai/xlnet · +24 2019
2 XLNet+DSC 89.5187.65 Dice Loss for Data-imbalanced NLP Tasks ShannonAI/dice_loss_for_NLP · fursovia/self-adj-dice · MindCode-4/code-6 · +1 2019
3 RoBERTa (no data aug) 89.486.5 ✓ RoBERTa: A Robustly Optimized BERT Pretraining Approach huggingface/transformers · pytorch/fairseq · PaddlePaddle/PaddleNLP · +64 2019
4 ALBERT xxlarge 88.185.1 ALBERT: A Lite BERT for Self-supervised Learning of Language Representations huggingface/transformers · tensorflow/models · PaddlePaddle/PaddleNLP · +45 2019
5 SG-Net 87.985.1 SG-Net: Syntax-Guided Machine Reading Comprehension cooelf/SG-Net 2019
6 SpanBERT 86.8– SpanBERT: Improving Pre-training by Representing and Predicting Spans facebookresearch/SpanBERT · mandarjoshi90/coref · zixinzeng-jennifer/spanbert_trans · +3 2019
7 ALBERT xlarge 85.983.1 ALBERT: A Lite BERT for Self-supervised Learning of Language Representations huggingface/transformers · tensorflow/models · PaddlePaddle/PaddleNLP · +45 2019
8 SemBERT large 83.680.9 Semantics-aware BERT for Language Understanding cooelf/SemBERT 2019
9 ALBERT large 82.179.0 ALBERT: A Lite BERT for Self-supervised Learning of Language Representations huggingface/transformers · tensorflow/models · PaddlePaddle/PaddleNLP · +45 2019
10 ALBERT base 79.176.1 ALBERT: A Lite BERT for Self-supervised Learning of Language Representations huggingface/transformers · tensorflow/models · PaddlePaddle/PaddleNLP · +45 2019
11 RMR + ELMo (Model-III) 74.872.3 Read + Verify: Machine Reading Comprehension with Unanswerable Questions 2018
12 U-Net 74.0 70.3 U-Net: Machine Reading Comprehension with Unanswerable Questions FudanNLP/UNet 2018
13 TinyBERT-6 67M 73.469.9 TinyBERT: Distilling BERT for Natural Language Understanding PaddlePaddle/PaddleNLP · huawei-noah/Pretrained-Language-Model · mindspore-ai/models · +7 2019
1–13 / 13 페이지당 10 20 50 100