What Ai Images Reveal About Our World

«Евразийская», финал, 2024/25, классы: 11, 10

Раздел «Аудирование» состоит из 5 заданий (1) – 5). Прослушайте аудиозапись дважды и выполните задания, выбрав необходимый вариант ответа в каждом пункте. Укажите номера выбранных вариантов в талоне ответов в строчке, соответствующей номеру задания, под буквой соответствующего пункта.

Task 1. По фрагментам аудиозаписи определите, о чем идет речь.

A. too advanced to be real

1 categorized and annotated datasets 2 inputting tags and language to produce images 3 the repetitive “clickwork” 4 the complexity of a basic phrase and even the simplest image 5 the idea of an AI artist software 6 text-generating platform Dall-E 7 first attempt at image generation and its challenges 8 late 20th century optical character recognition 9 identified patterns of pixels 10 the architecture and maintenance of text-to-image tools

B. to convert PDFs

1 categorized and annotated datasets 2 inputting tags and language to produce images 3 the repetitive “clickwork” 4 the complexity of a basic phrase and even the simplest image 5 the idea of an AI artist software 6 text-generating platform Dall-E 7 first attempt at image generation and its challenges 8 late 20th century optical character recognition 9 identified patterns of pixels 10 the architecture and maintenance of text-to-image tools

C. these operations

1 categorized and annotated datasets 2 inputting tags and language to produce images 3 the repetitive “clickwork” 4 the complexity of a basic phrase and even the simplest image 5 the idea of an AI artist software 6 text-generating platform Dall-E 7 first attempt at image generation and its challenges 8 late 20th century optical character recognition 9 identified patterns of pixels 10 the architecture and maintenance of text-to-image tools

D. its subject matter

1 categorized and annotated datasets 2 inputting tags and language to produce images 3 the repetitive “clickwork” 4 the complexity of a basic phrase and even the simplest image 5 the idea of an AI artist software 6 text-generating platform Dall-E 7 first attempt at image generation and its challenges 8 late 20th century optical character recognition 9 identified patterns of pixels 10 the architecture and maintenance of text-to-image tools

E. all the way down

1 categorized and annotated datasets 2 inputting tags and language to produce images 3 the repetitive “clickwork” 4 the complexity of a basic phrase and even the simplest image 5 the idea of an AI artist software 6 text-generating platform Dall-E 7 first attempt at image generation and its challenges 8 late 20th century optical character recognition 9 identified patterns of pixels 10 the architecture and maintenance of text-to-image tools

Task 2. Прочитайте приведенные ниже утверждения (1) – 10) и определите пять утверждений, информации о которых нет в тексте аудиозаписи.

The hype around the “AI artist” came as a bit of a shock since there hadn’t been various successful launches of such programs before. Back in the1990s, neural networks were able to process typed and handwritten characters. Optical character recognition outputs human language rather than lines of code. Natural language processing involves both applied linguistics and statistics. Patterns of pixels are assigned respective notions in the language to be later used in voice recognition systems. Researchers had to face the fact that image-to-text processing is more complex. In order to achieve good text generation quality, twice as much data need to be collected by the neural network. The authors of Generative Adversarial Text to Image Synthesis uploaded random datasets to train their model. The Caltech-UCSD Birds dataset was manually annotated by hired employees. The amount of human involvement needed to maintain neural networks opens a competitive and well-paid niche in the labour market.

Task 3. Выберите контекстуальные синонимы к словам (a) – e), употреблённым в тексте аудиозаписи.

a) a precursor

1 a prerequisite
2 an opponent
3 a forerunner
4 an indicator

b) a caption

1 a blueprint
2 an inscription
3 a pursuit
4 a caution

c) to decipher

1 to fabricate
2 to coincide
3 to fine-tune
4 to unravel

d) a snippet

1 a shred
2 a gauge
3 a puppet
4 a sneak

e) quotidian

1 partial
2 commonplace
3 urban
4 recognizable

Task 4. Выберите верное толкование выражений (a) – e), употреблённых в тексте аудиозаписи.

a) come down the pipeline

1 unfit to be manufactured
2 to disclose confidential information
3 to go unnoticed
4 to be heading our way

b) in fits and starts

1 half-baked
2 done sporadically
3 smoothly running
4 logically flawed

c) to flip the order

1 to confirm the commission
2 to overthrow the authority
3 to do the reverse
4 to right the wrong

d) less than straightforward

1 ambiguous
2 clear-cut
3 nonchalant
4 dispassionate

e) workers paid pennies

1 progressively higher wages
2 payment in advance
3 fundraising
4 via small transactions

Task 5. Завершите фразы, выбрав один из предложенных вариантов. Внимание! Варианты не могут повторяться.

a) The speaker uses ‘big news’

1 to refer to the same notion using a word from a different register
2 to build up anticipation
3 to give the reason why it was referred to as such
4 to imply the opposite of what they are saying
5 to emphasize the statement
6 to express their concern
7 to demonstrate what has been achieved
8 to avoid redundancy
9 to conclude in an ironic way
10 to refer to a common misconception

b) The speaker uses ‘editable’

1 to refer to the same notion using a word from a different register
2 to build up anticipation
3 to give the reason why it was referred to as such
4 to imply the opposite of what they are saying
5 to emphasize the statement
6 to express their concern
7 to demonstrate what has been achieved
8 to avoid redundancy
9 to conclude in an ironic way
10 to refer to a common misconception

c) The speaker uses ‘feline’

1 to refer to the same notion using a word from a different register
2 to build up anticipation
3 to give the reason why it was referred to as such
4 to imply the opposite of what they are saying
5 to emphasize the statement
6 to express their concern
7 to demonstrate what has been achieved
8 to avoid redundancy
9 to conclude in an ironic way
10 to refer to a common misconception

d) The speaker uses ‘pivotal’

1 to refer to the same notion using a word from a different register
2 to build up anticipation
3 to give the reason why it was referred to as such
4 to imply the opposite of what they are saying
5 to emphasize the statement
6 to express their concern
7 to demonstrate what has been achieved
8 to avoid redundancy
9 to conclude in an ironic way
10 to refer to a common misconception

e) The speaker uses ‘artificial artificial intelligence’

1 to refer to the same notion using a word from a different register
2 to build up anticipation
3 to give the reason why it was referred to as such
4 to imply the opposite of what they are saying
5 to emphasize the statement
6 to express their concern
7 to demonstrate what has been achieved
8 to avoid redundancy
9 to conclude in an ironic way
10 to refer to a common misconception

User Details

Name
Email

Prove you aren't a robot. How many letters 'a' is there in the word 'cat'? (e.g: 3)

Евразийская олимпиада школьников — перечневая олимпиада, обычно I или II уровня, которую проводит МГЛУ совместно с несколькими другими университетами России. Почитать об особенностях этой олимпиады и посмотреть на примеры заданий, которые в ней появлялись, можно на странице Евразийской олимпиады сайта Upwego. Также можно посетить официальную страницу олимпиады.

Чтобы получить диплом Евразийской олимпиады, нужно не только отлично знать английский язык, но и обладать широким кругозором — как общекультурным, так и лингвистическим, поскольку задания этой олимпиады разносторонние и творческие. На занятиях в мини-группах можно ознакомиться с разнообразными лингвистическими особенностями и потренировать как навыки письма, так и говорения (устный тур на этой олимпиаде тоже есть!), а олимпиадный марафон Upwego поможет построить фундамент подготовки ко всем остальным разделам олимпиады :)

artificial intelligence | Евразийская | Евразийская финал 11 | art | technology