Survey about Works of MOSAIC
MOSAIC 是 AI2 (ALLEN Institute for Artificial Intelligence) 研究院中进行机器常识智能研究的小组[1].
现有的研究项目：(Update at July, 2019)
Commonsense Knowledge Graphs: Explori
...
20190121 | 清华大学AI研究院知识智能研究中心发布会的记录
Minute and thoughts of the launch conference of THUKC (清华大学人工智能研究院知识智能研究中心).Mainly about the part of Commonsense-Awared Natural Language Generation (常
...
Natural Language Inference | Analysis and Paper List
NLI IntroductionChallenges
词语匹配的多元性（同义词、一词多义）lexical gap、lexical variant；
短语匹配的结构性、多粒度计算语义；
文本匹配的层次性；
利用不相似部分的信息；
自然语言的基础现象：Variability of Semantic ex
...
WIP| Analysis of FEVER
FEVER Introduction
FEVER data: link
包含 185,445 条断言(claims)
FEVER shared Task(with NAACL 2018) info: details here
FEVER official baseline code: githu
...
MRC 的评测与所需能力
Reading Comprehension Skills
An Analysis of Prerequisite Skills for Reading Comprehension.[1]Prerequisite skills for reading comprehension: Multi-pers
...
机器阅读理解数据集
Contributed by Luxi Xing and Yuqiang Xie. IIE, CAS.
IntroductionBased on the style of the ANSWER for each datasets, we split the datasets into several
...
Multi-Granular Sequence Encoding via Dilated Compositional Units for Reading Comprehension
Multi-Granular Sequence Encoding via Dilated Compositional Units for Reading ComprehensionEMNLP 2018Yi Tay et al.多粒度/尺度序列编码; 膨胀组合单元
Motivation
Seque
...
WIP| On the Capabilities and Limitations of Reasoning for Natural Language Understanding
[Under reading]
On the Capabilities and Limitations of Reasoning for Natural Language UnderstandingKhashabi et al.作者来自 University of Pennsylvania, In
...
Text Summarization - Main Problems
文本摘要中的主要问题与挑战
问题：区分是挑战还是技术问题
1.关注重点的词、句；
Attention、pos、tf-idf
intra-temporal attention
encoder-less self-attention 【generating Wikipedia by Summarizin
...
Neural Machine Translation - Main Problems
神经机器翻译中的主要问题与挑战
1.难以跨领域神经机器翻译（NMT）在处理领域之外的数据时的表现很糟：当前的机器翻译系统会生成非常流畅的输出，这些输出与领域外数据的输入无关。因此像Google翻译这样的通用机器翻译系统在法律或金融等专业领域的表现尤其糟糕。与基于短语的系统等传统方法相比，NMT系统的
...