Test 10 · What is AI? Pattern-guessers, not people

Sam's camera made his face look smooth in a photo. What should he remember?

Mova's choice: option 4 · The right answer: option 4 · right ✓ · Who acts: camera, face · What happens: make, look, remember · To what, where: face

This is test 10 of lesson 1 «What is AI? Pattern-guessers, not people» of the AI course of Mova's primary school. The page gives the question's syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause, then the 4 options as the student saw them, each with its tree and pragmatics. Every text is parsed by Mova itself. How the student answered it, with the score of every option and why: the test at school.

The question

Sentence 1: Sam's camera made his face look smooth in a photo.

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Sam	Sam	PROPN	Number=Sing	3 camera	nmod:poss
2	's	's	PART	_	1 Sam	case
3	camera	camera	NOUN	Number=Sing	4 made	nsubj
4	made	make	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	root	root
5	his	his	PRON	Case=Gen Gender=Masc Number=Sing Person=3 Poss=Yes PronType=Prs	6 face	nmod:poss
6	face	face	NOUN	Number=Sing	4 made	obj
7	look	look	VERB	VerbForm=Inf	4 made	xcomp
8	smooth	smooth	ADJ	Degree=Pos	7 look	xcomp
9	in	in	ADP	_	11 photo	case
10	a	a	DET	Definite=Ind PronType=Art	11 photo	det
11	photo	photo	NOUN	Number=Sing	8 smooth	obl
12	.	.	PUNCT	_	4 made	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	make	past	Sam's camera	camera's face					tree rules
2	look	infinitive	camera's face not said here					smooth property	vn stimulus_subject-30.4 «NP V ADJ»: agent=Stimulus

One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.

Sentence 2: What should he remember?

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	What	what	PRON	PronType=Int	4 remember	obj
2	should	should	AUX	VerbForm=Fin	4 remember	aux
3	he	he	PRON	Case=Nom Gender=Masc Number=Sing Person=3 PronType=Prs	4 remember	nsubj
4	remember	remember	VERB	VerbForm=Inf	root	root
5	?	?	PUNCT	_	4 remember	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	remember	should	camera						vn characterize-29.2-1-1 «NP V NP S_ING»: agent=Agent what=Theme

The options

Option 1

His face really is that smooth in real life, the camera just shows it

The tree, 14 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	His	his	PRON	Case=Gen Gender=Masc Number=Sing Person=3 Poss=Yes PronType=Prs	2 face	nmod:poss
2	face	face	NOUN	Number=Sing	6 smooth	nsubj
3	really	really	ADV	_	6 smooth	advmod
4	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	6 smooth	cop
5	that	that	SCONJ	_	6 smooth	mark
6	smooth	smooth	ADJ	Degree=Pos	root	root
7	in	in	ADP	_	9 life	case
8	real	real	ADJ	Degree=Pos	9 life	amod
9	life	life	NOUN	Number=Sing	6 smooth	obl
10	,	,	PUNCT	_	14 shows	punct
11	the	the	DET	Definite=Def PronType=Art	12 camera	det
12	camera	camera	NOUN	Number=Sing	14 shows	nsubj
13	just	just	ADV	_	14 shows	advmod
14	shows	show	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	6 smooth	conj
15	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	14 shows	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	is	present	His face			real life		smooth property	tree rules
2	show	present	the camera	face					vn transfer_mesg-37.1.1-1-1 «NP V NP.recipient»: agent=Agent it=Recipient

Option 2

The camera can see things eyes cannot

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	The	the	DET	Definite=Def PronType=Art	2 camera	det
2	camera	camera	NOUN	Number=Sing	4 see	nsubj
3	can	can	AUX	VerbForm=Fin	4 see	aux
4	see	see	VERB	VerbForm=Inf	root	root
5	things	thing	NOUN	Number=Plur	4 see	obj
6	eyes	eye	NOUN	Number=Plur	7 can	nsubj
7	can	can	AUX	VerbForm=Fin	5 things	acl:relcl
8	not	not	PART	Polarity=Neg	4 see	advmod

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	not see	can	the camera	things					vn comprehend-87.2-1-1 «NP V NP»: agent=Experiencer thing=Stimulus

Option 3

Smoothed photos are always more truthful

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Smoothed	smooth	VERB	Tense=Past VerbForm=Part Voice=Pass	2 photos	amod
2	photos	photo	NOUN	Number=Plur	6 truthful	nsubj
3	are	be	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	6 truthful	cop
4	always	always	ADV	PronType=Tot	5 more	advmod
5	more	more	ADV	Degree=Cmp	6 truthful	advmod
6	truthful	truthful	ADJ	Degree=Pos	root	root

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	smooth	participle							vn other_cos-45.4 «NP.patient V»: agent=Patient
2	is	present	Smoothed photos					truthful property	tree rules

Option 4 · ← Mova's choice · ✓ the right answer

Photos can be quietly changed, so real faces don't look like that

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Photos	photo	NOUN	Number=Plur	5 changed	nsubj:pass
2	can	can	AUX	VerbForm=Fin	5 changed	aux
3	be	be	AUX	VerbForm=Inf	5 changed	aux:pass
4	quietly	quietly	ADV	_	5 changed	advmod
5	changed	change	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
6	,	,	PUNCT	_	12 look	punct
7	so	so	ADV	_	12 look	advmod
8	real	real	ADJ	Degree=Pos	9 faces	amod
9	faces	face	NOUN	Number=Plur	12 look	nsubj
10	do	do	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	12 look	aux
11	n't	not	PART	Polarity=Neg	12 look	advmod
12	look	look	VERB	VerbForm=Inf	5 changed	parataxis
13	like	like	ADP	_	14 that	case
14	that	that	PRON	Number=Sing PronType=Dem	12 look	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	change	can · passive		Photos				quietly property	vn dress-41.1.1-1-1 «NP V NP»: subject=Patient
2	not look	present · do-support	real faces					like that	vn stimulus_subject-30.4 «NP V PP.attribute»: agent=Stimulus

Text: the tests of Mova's primary school, written for Mova (world/data/school/), as the course runner gave them to the student. Trees: Mova's own parse — the en model on the CPU, when the site is built — not checked by a teacher; they may be wrong. Frames: Mova's frames reader (prag), each text read on its own. Elamur Cosmic License.

This page in Markdown → https://md.mova.elamur.ai/free/en/corpora/questions/ai/1-what-is-ai/t10/