OpenCodePapers

question-answering-on-squad20-dev

Question Answering
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeF1EMModelNameReleaseDate
XLNet: Generalized Autoregressive Pretraining for Language Understanding✓ Link90.687.9XLNet (single model)2019-06-19
Dice Loss for Data-imbalanced NLP Tasks✓ Link89.5187.65XLNet+DSC2019-11-07
RoBERTa: A Robustly Optimized BERT Pretraining Approach✓ Link89.486.5RoBERTa (no data aug)2019-07-26
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations✓ Link88.185.1ALBERT xxlarge2019-09-26
SG-Net: Syntax-Guided Machine Reading Comprehension✓ Link87.985.1SG-Net2019-08-14
SpanBERT: Improving Pre-training by Representing and Predicting Spans✓ Link86.8SpanBERT2019-07-24
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations✓ Link85.983.1ALBERT xlarge2019-09-26
Semantics-aware BERT for Language Understanding✓ Link83.680.9SemBERT large2019-09-05
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations✓ Link82.179.0ALBERT large2019-09-26
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations✓ Link79.176.1ALBERT base2019-09-26
Read + Verify: Machine Reading Comprehension with Unanswerable Questions74.872.3RMR + ELMo (Model-III)2018-08-17
U-Net: Machine Reading Comprehension with Unanswerable Questions✓ Link74.0 70.3U-Net2018-10-12
TinyBERT: Distilling BERT for Natural Language Understanding✓ Link73.469.9TinyBERT-6 67M2019-09-23
Check out how to contribute new results. Then edit this file.