推荐daddyo · 2017-07
“Learning to Compose Neural Networks for Question Answering (Learning to Compose Neural Networks for Question Answering) comes close to breaking this barrier.”
认为这篇论文接近突破这道障碍
按问题动态拼装模块化神经网络,用于图像与知识库问答
提出一种同时适用于图像和结构化知识库的问答模型:根据自然语言问题,从一组可组合的模块中自动拼装出神经网络。模块参数与网络拼装参数通过强化学习联合学习,只需(场景、问题、答案)三元组作为监督。该动态神经模块网络在视觉和结构化两类基准上都取得了当时最好的结果。
评论者把它列为让神经网络按问题组合模块、做组合式问答的有趣工作,并收入自己推荐的 NLP 论文清单。
“Learning to Compose Neural Networks for Question Answering (Learning to Compose Neural Networks for Question Answering) comes close to breaking this barrier.”
“Edit: Adding a few interesting older NLP papers that I came across this year.”