Exam question 3 · Stories, jokes and riddles with a chatbot

Identify the principal reason for which a chatbot reproduces well-known jokes in preference to novel ones.

Mova's choice: option 3 · The right answer: option 3 · right ✓ · Who acts: you, chatbot · What happens: identify, reproduce, know · To what, where: principal reason, known jokes, novel ones, preference

This is exam question 3 of lesson 4 «Stories, jokes and riddles with a chatbot» 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	Identify	identify	VERB	Mood=Imp VerbForm=Fin	root	root
2	the	the	DET	Definite=Def PronType=Art	4 reason	det
3	principal	principal	ADJ	Degree=Pos	4 reason	amod
4	reason	reason	NOUN	Number=Sing	1 Identify	obj
5	for	for	ADP	_	6 which	case
6	which	which	PRON	PronType=Rel	9 reproduces	obl
7	a	a	DET	Definite=Ind PronType=Art	8 chatbot	det
8	chatbot	chatbot	NOUN	Number=Sing	9 reproduces	nsubj
9	reproduces	reproduce	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	4 reason	parataxis
10	well	well	ADV	Degree=Pos	12 known	advmod
11	-	-	PUNCT	_	10 well	punct
12	known	know	VERB	Tense=Past VerbForm=Part Voice=Pass	13 jokes	amod
13	jokes	joke	NOUN	Number=Plur	9 reproduces	obj
14	in	in	ADP	_	15 preference	case
15	preference	preference	NOUN	Number=Sing	9 reproduces	obl
16	to	to	ADP	_	18 ones	case
17	novel	novel	ADJ	Degree=Pos	18 ones	amod
18	ones	one	NOUN	Number=Plur	9 reproduces	obl
19	.	.	PUNCT	_	1 Identify	punct

#	verb	tense, mood	who	what, whom	to whom	where	other	roles by
1	identify	imperative	you not said here	the principal reason				vn characterize-29.2-1-1 «NP V NP S_ING»: agent=Agent reason=Theme
2	reproduce	present	a chatbot	known jokes	novel ones	preference	for the principal reason	vn other_cos-45.4 «NP V NP.patient»: agent=Agent joke=Patient
3	know	participle					well property	vn comprehend-87.2-1-1-1 «NP V»: agent=Experiencer

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

It is incapable of producing any sentence not seen before

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 incapable	nsubj
2	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	3 incapable	cop
3	incapable	incapable	ADJ	Degree=Pos	root	root
4	of	of	SCONJ	_	5 producing	mark
5	producing	produce	VERB	VerbForm=Ger	3 incapable	advcl
6	any	any	DET	PronType=Ind	7 sentence	det
7	sentence	sentence	NOUN	Number=Sing	5 producing	obj
8	not	not	PART	Polarity=Neg	9 seen	advmod
9	seen	see	VERB	Tense=Past VerbForm=Part Voice=Pass	5 producing	advcl
10	before	before	ADV	_	9 seen	advmod

#	verb	tense, mood	who	what, whom	other	roles by
1	is	present	It		incapable property	tree rules
2	produce	gerund		any sentence		vn create-26.4 «NP V NP»: agent=Agent sentence=Result
3	not see	participle			before property	vn comprehend-87.2-1-1 «NP V»: agent=Experiencer

Option 2

It is required by its makers to repeat a single joke

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 required	nsubj:pass
2	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	3 required	aux:pass
3	required	require	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
4	by	by	ADP	_	6 makers	case
5	its	its	PRON	Case=Gen Gender=Neut Number=Sing Person=3 Poss=Yes PronType=Prs	6 makers	nmod:poss
6	makers	maker	NOUN	Number=Plur	3 required	obl
7	to	to	PART	_	8 repeat	mark
8	repeat	repeat	VERB	VerbForm=Inf	3 required	xcomp
9	a	a	DET	Definite=Ind PronType=Art	11 joke	det
10	single	single	ADJ	Degree=Pos	11 joke	amod
11	joke	joke	NOUN	Number=Sing	8 repeat	obj

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	require	present · passive	its makers	It					vn order-58.3-1 «NP V NP S_ING»: agent=Agent subject=Recipient clause=Topic
2	repeat	to-infinitive	its makers not said here	a single joke					vn begin-55.1-1-1 «NP V NP»: agent=Agent joke=Eventuality

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

Such jokes occur most frequently in the text from which it learned

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	Such	such	ADJ	Degree=Pos	2 jokes	amod
2	jokes	joke	NOUN	Number=Plur	3 occur	nsubj
3	occur	occur	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	root	root
4	most	most	ADV	Degree=Sup	5 frequently	advmod
5	frequently	frequently	ADV	_	3 occur	advmod
6	in	in	ADP	_	8 text	case
7	the	the	DET	Definite=Def PronType=Art	8 text	det
8	text	text	NOUN	Number=Sing	3 occur	obl
9	from	from	ADP	_	10 which	case
10	which	which	PRON	PronType=Rel	12 learned	obl
11	it	it	PRON	Case=Nom Gender=Neut Number=Sing Person=3 PronType=Prs	12 learned	nsubj
12	learned	learn	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	8 text	acl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	occur	present	Such jokes			the text		frequently property	vn occur-48.3-1 «NP V PP»: agent=Theme in text=Location
2	learn	past	text			the text			vn learn-14-2-1 «NP V PP.source»: agent=Agent from which=Source

Option 4

It regards novel jokes as unsuitable for children

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 regards	nsubj
2	regards	regard	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	root	root
3	novel	novel	ADJ	Degree=Pos	4 jokes	amod
4	jokes	joke	NOUN	Number=Plur	2 regards	obj
5	as	as	ADP	_	6 unsuitable	case
6	unsuitable	unsuitable	ADJ	Degree=Pos	2 regards	obl
7	for	for	ADP	_	8 children	case
8	children	child	NOUN	Number=Plur	2 regards	obl

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	regard	present	It	novel jokes				as unsuitable; for children	vn assessment-34.1 «NP V NP PP.attribute»: agent=Agent joke=Theme for child=Attribute

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/4-stories-jokes-and-riddles/e3/