Question

    An artificial intelligence can accurately translate thoughts into sentences, at least for a limited vocabulary of 250 words. The system may bring us a step closer to _______ speech to people who have lost the ability.

Joseph Makin at the University of California, San Francisco, and his colleagues used deep learning algorithms(算法) to study the brain _______ of four women as they spoke. The women, who all suffer from a certain kind of brain disorder, already had electrodes attached to their brains to monitor disease attacks.

Each woman was asked to read aloud from a set of sentences as the team _______ brain activity. The largest group of sentences _______ 250 unique words. The team fed this brain activity to a network algorithm related to nerves, training it to _______ regularly occurring patterns that could be linked to repeated aspects of speech. These patterns were then fed to a second network, which tried to turn them into words to _________ a sentence.

Each woman repeated the sentences at least twice, and the final repetition didn’t form part of the training data, _______ the researchers to test the system.Each time a person speaks the same sentence, the brain activity associated will be similar but not exactly the same. “Memorizing the brain activity of these sentences wouldn’t help, _______ the network instead has to learn what’s similar about them so that it can generalize to this final example,” says Makin. Across the four women, the AI’s best performance was an average translation error rate of 3 per cent.

Makin says that using a small number of sentences made it _______ for the AI to learn which words tend to follow others. For example, the AI was able to __________ that the word “Bear” was always likely to follow the word “Teddy” in a certain set of sentences, from brain activity alone.

The team tried transforming the brain signal data into __________ words at a time, rather than whole sentences, but this __________ the error rate to 38 per cent even for the best performance. “So the network clearly is learning facts about which words go together, and not just which brain activity __________ with which words,”says Makin.

This will make it hard to scale up the system to a/an __________ vocabulary because each new word increases the number of possible sentences, reducing __________. Sophie Scott at University College London says we are a long way from being able to translate brain signal data comprehensively.

1.A.assigning B.conveying C.restoring D.introducing

2.A.systems B.signals C.signatures D.symbols

3.A.illuminated B.discovered C.measured D.stopped

4.A.consisted of B.adjusted to C.agreed with D.focused on

5.A.simplify B.identify C.intensify D.justify

6.A.understand B.form C.describe D.judge

7.A.allowing B.inspiring C.instructing D.advising

8.A.because B.so C.if D.but

9.A.quicker B.slower C.easier D.tougher

10.A.split B.reflect C.decode D.tear

11.A.individual B.common C.modified D.technical

12.A.increased B.decreased C.leveled D.degraded

13.A.furnished B.mixed C.associated D.armed

14.A.passive B.active C.limited D.expanded

15.A.tendency B.currency C.accuracy D.fluency

Answer

1.C

2.B

3.C

4.A

5.B

6.B

7.A

8.B

9.C

10.C

11.A

12.A

13.C

14.D

15.C

【解析】

这是一篇说明文。文章主要描述了一项关于人工智能可以将人类大脑的信号转化成词汇、句子的研究。文章花了大篇幅来描述研究的过程。最终表明将人类大脑的信号彻底的翻译出来还需一段很长的路要走。

1.考查动词词义辨析。句意：这个系统让我们更进一步去重建那些失去说话功能的人的语言能力。A. 分配、分发；B. 传达、传递；C. 恢复、重建；D. 介绍、引进。根据前句An artificial intelligence can accurately translate thoughts into sentences, at least for a limited vocabulary of 250 words可知，人工智能能够精确将想法翻译成句，虽有字数限制。因此推断它有助于恢复那些失去说话能力的人讲话能力。故选C项。

2.考查名词词义辨析。句意：加利福利亚大学的Joseph Makin和他的同事运用深入学习算法来研究4个女性说话时的脑部信号。A. 系统；B. 信号；C. 签名、鲜明特征；D. 符号、象征。根据后文第11题The team tried transforming the brain signal data into…..可知，他们对4个女性在讲话期间的脑部信号进行研究。故选B项。

3.考查动词词义辨析。句意：当这个团队在对他们的脑部活动进行测量之时，每个女性被要求大声朗读一组句子。A. 照亮、阐明、启发；B. 发现；C. 衡量、丈量、测量；D. 停止、阻止。根据上文The women, who all suffer from a certain kind of brain disorder, already had electrodes attached to their brains to monitor disease attacks可知，每个实验女性的脑部都安装了一个电极来监视脑部疾病的发作。故可以了解这个装置也可用来测量脑部的活动。故选C项。

4.考查动词短语辨析。句意：最大的句群包括了250个独一无二的单词。A. 由……组成；B. 调整、调节；C. 同意；适合；D. 专注于。句子由单个的词组成。故选A项。

5.考查动词词义辨析。句意：这个团队将脑部的活动注入了有关神经的算法网络中，训练这种算法去识别经常出现的与语言的重复方面有关的模块。A. 简化；B. 验证、识别；C. 使加强；D. 证明……合法。根据后文提及的AI能够破译一些词汇，能清楚的学习一些事实，可知这种算法可用来识别某些常见的模块。故选B。

6.考查动词词义辨析。句意：这些模块之后被注入第二个网络，这个网络尝试将他们变成词形成句子。A. 理解；B. 形成；C. 描述；D. 判断。根据前句The largest group of sentences consisted of 250 unique words可知，句子由词构成；多个词形成句子。故选B项。

7.考查动词词义辨析。句意：每个女性至少重复句子两次，之后的最后一次的重复不形成训练的数据的一部分，以允许研究者们测试系统。A. 允许；B. 鼓舞、激发；C. 指导；D. 建议。分析句意可知，最后的重复不构成训练数据的一部分，是为了让研究人员对系统进行测试。故选A。

8.考查连词词义辨析。句意：记忆这些句子的脑部活动是没有用的，所以这个网络不得不去了解他们之间的相似点是什么以至于他能将最后一个例子一般化。A. 因为；B. 所以；C. 如果、是否；D. 但是。根据语境，因为记忆脑部活动没有意义所以它才不得不去了解其它。故选B项。

9.考查形容词辨析及其比较级。句意：Makin说使用少量句子能够让AI更轻松的学到哪些词易于跟着其他词。A. 更快的；B. 更慢的；C.更轻松的；更简单的；D. 更艰难的。根据后句For example, the AI was able to decode that the word “Bear” was always likely to follow the word “Teddy” in a certain set of sentences, from brain activity alone可知，人工智有能力破译某些特定句子中的词和词的关系。故选C项。

10.考查动词词义辨析。句意：例如，单从大脑的活动而言，人工智能能够破译在某些特定的句中，“熊”这个词总是跟在“泰迪”后面。A. 分离、分开、劈开；B. 反映、反思；C. 破译；D. 撕毁。根据上文可知，人工智能能够轻松的知道某些词会比较容易放在一起的。而本句刚好是针对上句举的例子。和上句有关联。故选C项。

11.考查形容词辨析。句意：这个团队尝试，每次将大脑的信号数据转化为单个的词而不是整个句子。A. 独自的、单个的；B. 普通的；C. 改进的、修改的、改良的；D. 工艺的、科技的、技术上的。根据本句话的后半段rather than whole sentences可知，要填的词和后面的而不是的内容相反。后面讲整个句子，前面讲的应该是单个的词。故选A项。

12.考查动词词义辨析。句意：但是这次尽管是最好的表现，错误率却增加到了38%。A. 增加；B. 减少；C. 使同等、瞄准、拉平；D. 分解、削减、降低……身份。根据后文，每个新词都会增加可能句子的数量，从而减少精确性。减少精确性故会增加错误率。故选A项。

13.考查动词短语辨析。句意：Makin说“所以这个系统清楚的学到的事实是：哪些词是在一起的而不仅仅是哪一个脑部活动和哪些词有关”。A. furnish with供给，提供；B. mix with 与……混合；C. associate with和……有关；D. arm with 武装、装备。分析句子可知，网络学习的是哪些词在一起，而不是哪些大脑活动和哪些词有关。故选C项。

14.考查形容词词义辨析。这将使扩展系统以扩展词汇变得困难。A. 消极的、被动的；B. 积极地 、活跃的；C. 有限的；D. 扩大的，展开的。根据后句because each new word increases the number of possible sentences可知，新词增加了新句的可能性。故选D项。

15.考查名词词义辨析。句意：因为每个新词都会增加可能句子的数量，从而减少精确性。A. 倾向；B. 货币、通货；C. 精确性；D. 流畅性。根据后句Sophie Scott at University College London says we are a long way from being able to translate brain signal data comprehensively可知，彻底的将大脑信号翻译出来还有很长一段路要走。中间还有很多问题要解决。同时在第7题提到了：错误率有时候会达到38%。精确度不够。故选C项。