Mova0.5 pre-release⌕

☰

home / corpora / questions / AI / Olympiad: claims, clues and tricky mes… / t10

Test 10 · ★★★ · Olympiad: claims, clues and tricky messages ★★★

The airline bot "got it a tiny bit wrong" about refunds, a mate says. How big was the trouble really?

Mova's choice: option 1 · The right answer: option 4 · wrong ✗ · Who acts: airline bot, mate · What happens: get, say, be · To what, where: it, big trouble · Speech: someone speaks here

This is test 10 · ★★★ of lesson 32 «Olympiad: claims, clues and tricky messages ★★★» 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: The airline bot "got it a tiny bit wrong" about refunds, a mate says.

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	The	the	DET	Definite=Def PronType=Art	3 bot	det
2	airline	airline	NOUN	Number=Sing	3 bot	compound
3	bot	bot	NOUN	Number=Sing	5 got	nsubj
4	"	"	PUNCT	_	5 got	punct
5	got	get	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	17 says	ccomp
6	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	5 got	obj
7	a	a	DET	Definite=Ind PronType=Art	9 bit	det
8	tiny	tiny	ADJ	Degree=Pos	9 bit	amod
9	bit	bit	NOUN	Number=Sing	10 wrong	obl:unmarked
10	wrong	wrong	ADJ	Degree=Pos	5 got	conj
11	"	"	PUNCT	_	10 wrong	punct
12	about	about	ADP	_	13 refunds	case
13	refunds	refund	NOUN	Number=Plur	5 got	obl
14	,	,	PUNCT	_	17 says	punct
15	a	a	DET	Definite=Ind PronType=Art	16 mate	det
16	mate	mate	NOUN	Number=Sing	17 says	nsubj
17	says	say	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	root	root
18	.	.	PUNCT	_	17 says	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	get in a quote	past	the airline bot	it				about refunds	vn comprehend-87.2-1 «NP V NP»: agent=Experiencer it=Stimulus
2	say speech: assert	present	a mate						vn say-37.7-1-1 «NP V that S»: agent=Agent clause=Topic

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: How big was the trouble really?

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	How	how	ADV	PronType=Int	2 big	advmod
2	big	big	ADJ	Degree=Pos	5 trouble	amod
3	was	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	5 trouble	cop
4	the	the	DET	Definite=Def PronType=Art	5 trouble	det
5	trouble	trouble	NOUN	Number=Sing	root	root
6	really	really	ADV	_	5 trouble	advmod
7	?	?	PUNCT	_	5 trouble	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	is a	past		the big trouble					tree rules

The options

Option 1 · ← Mova's choice

So tiny that nobody noticed it at all

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	So	so	ADV	_	2 tiny	advmod
2	tiny	tiny	ADJ	Degree=Pos	root	root
3	that	that	SCONJ	_	5 noticed	mark
4	nobody	nobody	PRON	Number=Sing PronType=Neg	5 noticed	nsubj
5	noticed	notice	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	2 tiny	advcl
6	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	5 noticed	obj
7	at	at	ADP	_	8 all	case
8	all	all	DET	PronType=Tot	5 noticed	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	notice	past	nobody	it		all			vn see-30.1-1 «NP V S»: agent=Experiencer it=Stimulus

Option 2

So big that every airline had to stop flying

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	So	so	ADV	_	2 big	advmod
2	big	big	ADJ	Degree=Pos	root	root
3	that	that	SCONJ	_	6 had	mark
4	every	every	DET	PronType=Tot	5 airline	det
5	airline	airline	NOUN	Number=Sing	6 had	nsubj
6	had	have	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	2 big	advcl
7	to	to	PART	_	8 stop	mark
8	stop	stop	VERB	VerbForm=Inf	6 had	xcomp
9	flying	fly	VERB	VerbForm=Ger	8 stop	xcomp

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	have	past	every airline						tree rules
2	stop	to-infinitive	every airline not said here						vn stop-55.4-1-1 «NP V S_ING»: agent=Affector clause=Eventuality
3	fly	gerund	every airline not said here						vn spatial_configuration-47.6-1 «NP V»: agent=Theme

Option 3

Tiny, since bots can never be blamed for anything

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Tiny	tiny	ADJ	Degree=Pos	4 bots	amod
2	,	,	PUNCT	_	4 bots	punct
3	since	since	ADP	_	4 bots	case
4	bots	bot	NOUN	Number=Plur	8 blamed	nsubj:pass
5	can	can	AUX	VerbForm=Fin	8 blamed	aux
6	never	never	ADV	PronType=Neg	8 blamed	advmod
7	be	be	AUX	VerbForm=Inf	8 blamed	aux:pass
8	blamed	blame	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
9	for	for	ADP	_	10 anything	case
10	anything	anything	PRON	Number=Sing PronType=Ind	8 blamed	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	not blame	can · passive		Tiny bots				for anything	vn judgment-33.1 «NP V NP PP.attribute»: subject=Theme for anything=Attribute

Option 4 · ✓ the right answer

Big enough that a tribunal made the airline pay

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Big	big	ADJ	Degree=Pos	root	root
2	enough	enough	ADV	_	1 Big	advmod
3	that	that	SCONJ	_	6 made	mark
4	a	a	DET	Definite=Ind PronType=Art	5 tribunal	det
5	tribunal	tribunal	NOUN	Number=Sing	6 made	nsubj
6	made	make	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	1 Big	advcl
7	the	the	DET	Definite=Def PronType=Art	9 pay	det
8	airline	airline	ADJ	Degree=Pos	9 pay	amod
9	pay	pay	NOUN	Number=Sing	6 made	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	make	past	a tribunal	the airline pay					vn build-26.1-1 «NP.asset V NP»: agent=Asset pay=Product

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/32-olympiad-claims-clues-and-tricky-messages/t10/