Question Answering via Sentence Composition (QASC)

Aristo • 2019
QASC is a question-answering dataset with a focus on sentence composition. It consists of 9,980 8-way multiple-choice questions about grade school science (8,134 train, 926 dev, 920 test), and comes with a corpus of 17M sentences.

Leaderboard

Top Public Submissions
Details
Created
Accuracy
1
KF+SIR+2 Step
Pratyay Banerjee, Chitta Baral
12/7/201980%
2
Roberta-large; Two-Step(QASC)
Sun Yat-sen University
1/24/202080%
3
BERT-MCQ(BERT-LC[WM]; Two-step(QASC); RACE+SCI)
Aristo team at Allen Institute for AI
10/27/201973%
4
BERT-MCQ(BERT-LC; Two-step(QASC); RACE+SCI)
Aristo team at Allen Institute for AI
10/20/201968%
5
BERT-MCQ(BERT-LC; Two-step(QASC))
Aristo team at Allen Institute for AI
10/13/201967%