Exam question 3 · Using AI to learn, not to copy

Select the response that is consistent with the established facts concerning automated tools for the detection of machine-written text.

Mova's choice: option 3 · The right answer: option 3 · right ✓ · Who acts: you, response, established facts · What happens: select, be, establish, concern, write, detect · To what, where: response, automated tools, written text

This is exam question 3 of lesson 10 «Using AI to learn, not to copy» of the AI course of Mova's primary school, asked in stiff, formal language. 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

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Select	select	VERB	Mood=Imp VerbForm=Fin	root	root
2	the	the	DET	Definite=Def PronType=Art	3 response	det
3	response	response	NOUN	Number=Sing	1 Select	obj
4	that	that	PRON	PronType=Rel	6 consistent	nsubj
5	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	6 consistent	cop
6	consistent	consistent	ADJ	Degree=Pos	3 response	acl:relcl
7	with	with	ADP	_	10 facts	case
8	the	the	DET	Definite=Def PronType=Art	10 facts	det
9	established	establish	VERB	Tense=Past VerbForm=Part Voice=Pass	10 facts	amod
10	facts	fact	NOUN	Number=Plur	6 consistent	obl
11	concerning	concern	VERB	VerbForm=Ger	10 facts	acl
12	automated	automated	ADJ	Degree=Pos	13 tools	amod
13	tools	tool	NOUN	Number=Plur	11 concerning	obj
14	for	for	ADP	_	16 detection	case
15	the	the	DET	Definite=Def PronType=Art	16 detection	det
16	detection	detection	NOUN	Number=Sing	13 tools	nmod
17	of	of	ADP	_	21 text	case
18	machine	machine	NOUN	Number=Sing	20 written	compound
19	-	-	PUNCT	_	18 machine	punct
20	written	write	VERB	Tense=Past VerbForm=Part Voice=Pass	21 text	amod
21	text	text	NOUN	Number=Sing	16 detection	nmod
22	.	.	PUNCT	_	1 Select	punct

#	verb	tense, mood	who	what, whom	other	roles by
1	select	imperative	you not said here	the response		vn obtain-13.5.2 «NP V NP»: agent=Agent response=Theme
2	is	present	the response		with the established facts; consistent property	tree rules
3	establish	participle				vn indicate-78-1-1 «NP V NP to be NP»: agent=Pivot
4	concern	gerund	the established facts not said here	automated tools	for the detection of written text	vn amuse-31.1 «NP V NP»: agent=Stimulus tool=Experiencer
5	write	participle				vn scribble-25.2-1 «NP V»: agent=Agent
6	detect			written text		nom detection «cognition transitive»: of=patient

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.

The options

Option 1

They detect only errors of spelling

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	They	they	PRON	Case=Nom Number=Plur Person=3 PronType=Prs	2 detect	nsubj
2	detect	detect	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	root	root
3	only	only	ADV	_	4 errors	advmod
4	errors	error	NOUN	Number=Plur	2 detect	obj
5	of	of	ADP	_	6 spelling	case
6	spelling	spelling	NOUN	Number=Sing	4 errors	nmod

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	detect	present	They	errors of spelling					vn see-30.1-1 «NP V S»: agent=Experiencer error=Stimulus
2	err		spelling						nom error «act intransitive»: of=agent

Option 2

They are operated by the pupils themselves

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	They	they	PRON	Case=Nom Number=Plur Person=3 PronType=Prs	3 operated	nsubj
2	are	be	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	3 operated	aux:pass
3	operated	operate	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
4	by	by	ADP	_	6 pupils	case
5	the	the	DET	Definite=Def PronType=Art	6 pupils	det
6	pupils	pupil	NOUN	Number=Plur	3 operated	obl
7	themselves	themselves	PRON	Case=Acc Number=Plur Person=3 PronType=Prs Reflex=Yes	3 operated	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	operate	present · passive	the pupils	They				pupils = pupil obl	vn function-105.2.1-1 «NP V NP PP»: agent=Theme subject=Attribute

Option 3 · ← Mova's choice · ✓ the right answer

They are fallible and may wrongly accuse honest pupils

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	They	they	PRON	Case=Nom Number=Plur Person=3 PronType=Prs	3 fallible	nsubj
2	are	be	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	3 fallible	cop
3	fallible	fallible	ADJ	Degree=Pos	root	root
4	and	and	CCONJ	_	7 accuse	cc
5	may	may	AUX	VerbForm=Fin	7 accuse	aux
6	wrongly	wrongly	ADV	_	7 accuse	advmod
7	accuse	accuse	VERB	VerbForm=Inf	3 fallible	conj
8	honest	honest	ADJ	Degree=Pos	9 pupils	amod
9	pupils	pupil	NOUN	Number=Plur	7 accuse	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	is	present	They					fallible property	tree rules
2	accuse	may	They not said here	honest pupils				wrongly property	vn suspect-81 «NP V NP PP.attribute»: agent=Agent pupil=Theme

Option 4

They are infallible in every case

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	They	they	PRON	Case=Nom Number=Plur Person=3 PronType=Prs	3 infallible	nsubj
2	are	be	AUX	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	3 infallible	cop
3	infallible	infallible	ADJ	Degree=Pos	root	root
4	in	in	ADP	_	6 case	case
5	every	every	DET	PronType=Tot	6 case	det
6	case	case	NOUN	Number=Sing	3 infallible	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	is	present	They			every case		infallible property	tree rules

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/10-using-ai-to-learn-not-to-copy/e3/