HyperAI초신경

홈 플랫폼 문서 뉴스 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Question Answering On Cnn Daily Mail

평가 지표

CNN

Daily Mail

평가 결과

이 벤치마크에서 각 모델의 성능 결과

			Paper Title	Repository
GA+MAGE (32)	78.6	-	Linguistic Knowledge as Memory for Recurrent Neural Networks	-
GA Reader	77.9	80.9	Gated-Attention Readers for Text Comprehension
Attentive + relabling + ensemble	77.6	79.2	A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task
BiDAF	76.9	79.6	Bidirectional Attention Flow for Machine Comprehension
AIA	76.1	-	Iterative Alternating Neural Attention for Machine Reading
AS Reader (ensemble model)	75.4	77.7	Text Understanding with the Attention Sum Reader Network
ReasoNet	74.7	76.6	ReasoNet: Learning to Stop Reading in Machine Comprehension	-
AoA Reader	74.4	-	Attention-over-Attention Neural Networks for Reading Comprehension
EpiReader	74	-	Natural Language Comprehension with the EpiReader	-
Dynamic Entity Repres. + w2v	72.9	-	-	-
AttentiveReader + bilinear attention	72.4	75.8	A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task
AS Reader (single model)	69.5	73.9	Text Understanding with the Attention Sum Reader Network
MemNNs (ensemble)	69.4	-	Teaching Machines to Read and Comprehend
Classifier	67.9	68.3	A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task
Impatient Reader	63.8	68.0	Teaching Machines to Read and Comprehend
Attentive Reader	63	69	Teaching Machines to Read and Comprehend

0 of 16 row(s) selected.