Exam question 3 · Olympiad: thinking like a machine, Bebras puzzles ★★★

Which of the following statements most accurately characterises the error of the picture sorter that classified a husky upon snow as a wolf?

Mova's choice: option 1 · The right answer: option 2 · wrong ✗ · Who acts: picture sorter · What happens: follow, characteris, classify, err · To what, where: error, husky

This is exam question 3 of lesson 31 «Olympiad: thinking like a machine, Bebras puzzles ★★★» 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	Which	which	PRON	PronType=Rel	8 characterises	nsubj
2	of	of	ADP	_	5 statements	case
3	the	the	DET	Definite=Def PronType=Art	5 statements	det
4	following	follow	VERB	VerbForm=Ger	5 statements	amod
5	statements	statement	NOUN	Number=Plur	8 characterises	obl
6	most	most	ADV	Degree=Sup	7 accurately	advmod
7	accurately	accurately	ADV	_	8 characterises	advmod
8	characterises	characteris	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	root	root
9	the	the	DET	Definite=Def PronType=Art	10 error	det
10	error	error	NOUN	Number=Sing	8 characterises	obj
11	of	of	ADP	_	14 sorter	case
12	the	the	DET	Definite=Def PronType=Art	14 sorter	det
13	picture	picture	NOUN	Number=Sing	14 sorter	compound
14	sorter	sorter	NOUN	Number=Sing	10 error	nmod
15	that	that	PRON	PronType=Rel	16 classified	nsubj
16	classified	classify	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	14 sorter	acl:relcl
17	a	a	DET	Definite=Ind PronType=Art	18 husky	det
18	husky	husky	NOUN	Number=Sing	16 classified	obj
19	upon	upon	ADP	_	20 snow	case
20	snow	snow	NOUN	Number=Sing	18 husky	nmod
21	as	as	ADP	_	23 wolf	case
22	a	a	DET	Definite=Ind PronType=Art	23 wolf	det
23	wolf	wolf	NOUN	Number=Sing	16 classified	obl
24	?	?	PUNCT	_	8 characterises	punct

#	verb	tense, mood	who	what, whom	other	roles by
1	follow	gerund				vn occur-48.3-1 «NP V»: agent=Theme
2	characteris	present		the error of the picture sorter	of the following statements; accurately property	tree rules
3	classify	past	the picture sorter	a husky	upon snow; as a wolf	vn characterize-29.2 «NP V NP S_ING»: agent=Agent husky=Theme as wolf=Attribute
4	err		the picture sorter			nom error «act intransitive»: of=agent

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 · ← Mova's choice

It had been trained exclusively upon photographs of domestic cats

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	It	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	4 trained	nsubj:pass
2	had	have	AUX	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	4 trained	aux
3	been	be	AUX	Tense=Past VerbForm=Part	4 trained	aux:pass
4	trained	train	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
5	exclusively	exclusively	ADV	_	4 trained	advmod
6	upon	upon	ADP	_	7 photographs	case
7	photographs	photograph	NOUN	Number=Plur	4 trained	obl
8	of	of	ADP	_	10 cats	case
9	domestic	domestic	ADJ	Degree=Pos	10 cats	amod
10	cats	cat	NOUN	Number=Plur	7 photographs	nmod

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	train	past · perfect · passive		It				upon photographs of domestic cats; exclusively property	vn hire-13.5.3 «NP V NP»: subject=Theme

Option 2 · ✓ the right answer

It had associated the snowy background, rather than the animal, with wolves

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	It	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	3 associated	nsubj:pass
2	had	have	AUX	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	3 associated	aux
3	associated	associate	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
4	the	the	DET	Definite=Def PronType=Art	6 background	det
5	snowy	snowy	ADJ	Degree=Pos	6 background	amod
6	background	background	NOUN	Number=Sing	3 associated	obj
7	,	,	PUNCT	_	11 animal	punct
8	rather	rather	ADV	ExtPos=CCONJ	11 animal	cc
9	than	than	ADP	_	8 rather	fixed
10	the	the	DET	Definite=Def PronType=Art	11 animal	det
11	animal	animal	NOUN	Number=Sing	3 associated	conj
12	,	,	PUNCT	_	14 wolves	punct
13	with	with	ADP	_	14 wolves	case
14	wolves	wolf	NOUN	Number=Plur	3 associated	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	associate	past · perfect · passive		It				with wolves	tree rules

Option 3

It had been instructed deliberately to confuse the two species

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	It	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	4 instructed	nsubj:pass
2	had	have	AUX	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	4 instructed	aux
3	been	be	AUX	Tense=Past VerbForm=Part	4 instructed	aux:pass
4	instructed	instruct	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
5	deliberately	deliberately	ADV	_	4 instructed	advmod
6	to	to	PART	_	7 confuse	mark
7	confuse	confuse	VERB	VerbForm=Inf	4 instructed	advcl
8	the	the	DET	Definite=Def PronType=Art	10 species	det
9	two	two	NUM	NumForm=Word NumType=Card	10 species	nummod
10	species	species	NOUN	Number=Plur	7 confuse	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	instruct speech: other	past · perfect · passive		It				deliberately property	vn advise-37.9-1 «NP V NP.recipient»: subject=Recipient
2	confuse	to-infinitive		the two species					vn amalgamate-22.2-2 «NP V NP»: agent=Agent species=Patient

Option 4

It had insufficient light by which to perceive the photograph

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	It	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	2 had	nsubj
2	had	have	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	root	root
3	insufficient	insufficient	ADJ	Degree=Pos	4 light	amod
4	light	light	NOUN	Number=Sing	2 had	obj
5	by	by	ADP	_	6 which	case
6	which	which	PRON	PronType=Rel	2 had	obj
7	to	to	PART	_	8 perceive	mark
8	perceive	perceive	VERB	VerbForm=Inf	2 had	advcl
9	the	the	DET	Definite=Def PronType=Art	10 photograph	det
10	photograph	photograph	NOUN	Number=Sing	8 perceive	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	have	past	It	insufficient light					tree rules
2	perceive	to-infinitive		the photograph					vn see-30.1-1 «NP V S»: agent=Experiencer photograph=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/31-olympiad-thinking-like-a-machine/e3/