记录对比多个V-L数据集构建方法，主要包括：Vision-Text Retrieval、Visual Question Answering、Image Captioning、Visual Reasoning、Referring Expression、Visual Dialogue。

MaRVL多语Visual Language Reasoning数据集构建记录，参考论文Visually Grounded Reasoning across Languages and Cultures

图像语言预训练模型 (Vision-Language Pre-trained Model，VLPM) 总结。
主要参考论文Vision-and-Language Pretrained Models: A Survey和Trends in Integration of Vision and Language Research: A Survey of Tasks, Datasets, and Methods。

Wikidata的JSON版本数据结构以及一些相关的信息记录。

论文题目：A Survey on Knowledge Graphs: Representation, Acquisition and Applications

作者：Shaoxiong Ji, Shirui Pan, Erik Cambria, Pekka Marttinen, Philip S. Yu

单位：Aalto University, Monash University, Nanyang Technological University, University of Illinois

期刊：IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS

发表日期：2021.03.30

论文题目：X-FACTR: Multilingual Factual Knowledge Retrieval from Pretrained Language Models

作者：Zhengbao Jiang, Antonios Anastasopoulos, Jun Araki, Haibo Ding, Graham Neubig

单位：Carnegie Mellon University, George Mason University, Bosch Research

期刊：EMNLP 2020

发表日期：2020.10.13

跨语言预训练模型 (Cross-Lingual Pre-trained Language Model) 总结与对比。

知识增强预训练模型 (Knowledge Enhanced Pre-trained Language Model) 总结与对比。

2019年-2021年大规模预训练模型对比，从最开始2019年的8.5B的Megatron，到最近2021年底的280B的Gopher。

论文题目：ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation

作者：Yu Sun, Shuohuan Wang, Shikun Feng, Siyu Ding, Chao Pang, Junyuan Shang, Jiaxiang Liu, Xuyi Chen, Yanbin Zhao, Yuxiang Lu, Weixin Liu, Zhihua Wu, Weibao Gong, Jianzhong Liang, Zhizhou Shang, Peng Sun, Wei Liu, Xuan Ouyang, Dianhai Yu, Hao Tian, Hua Wu, Haifeng Wang

单位：Baidu Inc.

期刊：Arxiv

发表日期：2021.07.05