HyperAI超神经

首页算力平台文档资讯论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Question Answering On Babi

评估指标

Mean Error Rate

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
LSTM	28.7%	Scaling Memory-Augmented Neural Networks with Sparse Reads and Writes	-
EntNet	9.7%	Tracking the World State with Recurrent Entity Networks
End-To-End Memory Networks	7.5%	End-To-End Memory Networks
SDNC	6.4%	Scaling Memory-Augmented Neural Networks with Sparse Reads and Writes	-
NUTM	5.6%	Neural Stored-program Memory
ReMO	1.2%	Finding ReMO (Related Memory Object): A Simple Neural Architecture for Text based Reasoning	-
RR	0.46%	Recurrent Relational Networks
STM	0.39%	Self-Attentive Associative Memory
QRN	0.3%	Query-Reduction Networks for Question Answering
DMN+	-	Dynamic Neural Turing Machine with Soft and Hard Addressing Schemes	-
RUM	-	Rotational Unit of Memory
GORU	-	Gated Orthogonal Recurrent Units: On Learning to Forget
H-Mem	-	H-Mem: Harnessing synaptic plasticity with Hebbian Memory Networks	-
ours	-	Memory-enriched computation and learning in spiking neural networks through Hebbian plasticity

0 of 14 row(s) selected.

Question Answering On Babi | SOTA | HyperAI超神经