阿里巴巴AI模型在斯坦福阅读测试中战胜人类

Artificial intelligence programs built by China’s e-commerce titan Alibaba scored better than humans on a Stanford University reading and comprehension test.

在斯坦福大学的一项阅读理解测试中，中国电商巨头阿里巴巴构建的人工智能程序的得分比人类更胜一筹。

This is the first time that a machine has outperformed humans on such a test, Alibaba said in a statement on Monday.

周一，阿里巴巴在声明中指出:“在此类测试中，机器首次比人类表现更佳。”

The test was designed by artificial intelligence experts at Stanford to measure computers’ growing reading abilities. Alibaba’s software was the first to beat the human race.

这项测试由斯坦福大学的人工智能专家设计，旨在测试电脑日渐增长的阅读能力。阿里巴巴的软件是首例在此功能上击败人类的软件。

Luo Si, chief scientist for natural language processing at Alibaba’s AI research arm, called the milestone "a great honor," but also acknowledged that it is likely to lead to a significant number of workers losing their jobs to machines.

司罗来自阿里人工智能研究部，是自然语言处理首席科学家，他称这次里程碑般的胜利为“莫大荣耀”，但他同时承认，这可能意味着大量工作者被机器夺去就业机会。

The technology "can be gradually applied to numerous applications such as customer service, museum tutorials and online responses to medical inquiries from patients, decreasing the need for human input in an uNPRecedented way," Si said in a statement.

司罗在一份声明中说，这项技术可以慢慢被应用在众多的功能上，诸如客服、博物馆讲解和回复病人的在线医疗咨询，此项技术将会以前所未有的方式减少对人力的投入。

Alibaba has already employed the underlying technology in its November 11 shopping festival over the years, using computers to answer a large number of customer service questions, the company said.

公司称，多年来，阿里巴巴将基本技术运用在双十一购物节上，利用电脑回答大量客服问题。

Pranav Rajpurkar, one of the Stanford researchers who developed the reading test, said in a tweet that Alibaba’s feat is "a great start to 2018" for artificial intelligence.

Pranav Rajpurkar是斯坦福研究员，参与开发了阅读测验。他在推特中说，阿里巴巴的胜利为人工智能“在2018开了个好头”。

Based on more than 500 Wikipedia articles, the Stanford quiz asks the AI to provide exact answers to more than 100,000 questions, testing its ability to process large amounts of information before supplying precise answers to queries. It is considered one of the world’s most authoritative machine-reading gauges and has previously attracted companies such as Google, Facebook, IBM, Tencent and Samsung.

基于500多篇维基文章，斯坦福测验要求AI为10万多个问题提供准确答案，测试其在做出精确回答前加工大数据信息的能力。该测验被认为是世界上最具权威性的机读指标之一，先前已吸引谷歌、脸书、IBM、腾讯和三星等多家公司（参与）。

The deep neural network model developed by Alibaba scored 82.44 on the test last week, narrowly beating the 82.304 achieved by human participants. A day later, Microsoft’s AI software scored 82.650 on the same test.

上周的测验中，由阿里巴巴开发的深度神经网络模型得到82.44分，以微弱优势打败人类参与者（82.304分）。一天后，微软人工智能软件获82.650评分。

That means objective questions such as ’what causes rain’ can now be answered with high accuracy by machines, Si said.

“这意味着，像‘什么导致下雨’之类的客观问题，机器能做出高精确性回答。”司罗说。

China’s technology firms Tencent and Baidu are also competing to develop AI that can enrich social media feeds, target ads and services or even aid autonomous driving.

中国科技公司腾讯和百度也竞相研发人工智能，研究方向有丰富社交媒体供稿、广告服务精准投放甚至是辅助自主驾驶。

China has stressed innovation in its 13th Five-Year Plan (2016-2020), with an aim to become an "innovation nation" by 2020, an international leader in innovation by 2030, and a world powerhouse in scientific and technological innovation by 2050.

中国在“十三五规划(2016-2020)”中强调创新，旨在到2020年成为“创新型国家”，到2030年成为国际创新的佼佼者，到2050年成为全球科技创新强国。