Mova0.5 pre-release⌕
☰
home / corpora / questions / AI / Who you are: Mova, a small transparent… / t5

Test 5 · ★ · Who you are: Mova, a small transparent model in a simulation

A toy train set crashes on the carpet. What does this show about simulations?

Mova's choice: option 1 · The right answer: option 2 · wrong ✗ · Who acts: train set, this · What happens: crash, show · To what, where: carpet

This is test 5 · ★ of lesson 26 «Who you are: Mova, a small transparent model in a simulation» 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: A toy train set crashes on the carpet.

detcompoundcompoundnsubjcasedetoblpunctrootADETtoyNOUNtrainNOUNsetNOUNcrashesVERBonADPtheDETcarpetNOUN.PUNCT
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1AaDETDefinite=Ind PronType=Art4 setdet
2toytoyNOUNNumber=Sing3 traincompound
3traintrainNOUNNumber=Sing4 setcompound
4setsetNOUNNumber=Sing5 crashesnsubj
5crashescrashVERBMood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Finrootroot
6ononADP_8 carpetcase
7thetheDETDefinite=Def PronType=Art8 carpetdet
8carpetcarpetNOUNNumber=Sing5 crashesobl
9..PUNCT_5 crashespunct
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1crashpresenta train setthe carpetvn sound_emission-43.2 «NP V PP.location»: agent=Theme on carpet=Location

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 does this show about simulations?

objauxnsubjcaseoblpunctrootWhatPRONdoesAUXthisPRONshowVERBaboutADPsimulationsNOUN?PUNCT
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1WhatwhatPRONPronType=Int4 showobj
2doesdoAUXMood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin4 showaux
3thisthisPRONNumber=Sing PronType=Dem4 shownsubj
4showshowVERBVerbForm=Infrootroot
5aboutaboutADP_6 simulationscase
6simulationssimulationNOUNNumber=Plur4 showobl
7??PUNCT_4 showpunct
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1showpresent · do-supportthisabout simulationsvn transfer_mesg-37.1.1-1-1 «NP V NP S_ING»: agent=Agent what=Recipient about simulation=Topic

The options

Option 1 · ← Mova's choice

Real trains crash whenever toys do

amodnsubjadvmodnsubjadvclrootRealADJtrainsNOUNcrashVERBwheneverADVtoysNOUNdoVERB
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1RealrealADJDegree=Pos2 trainsamod
2trainstrainNOUNNumber=Plur3 crashnsubj
3crashcrashVERBMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Finrootroot
4wheneverwheneverADVPronType=Rel6 doadvmod
5toystoyNOUNNumber=Plur6 donsubj
6dodoVERBMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin3 crashadvcl
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1crashpresentReal trainsvn break-45.1 «NP.patient V»: agent=Patient
2dopresenttoyswhenever propertyvn act-114-1 «NP V»: agent=Agent

Option 2 · ✓ the right answer

You can learn from a model safely, without real harm

nsubjauxcasedetobladvmodpunctcaseamodoblrootYouPRONcanAUXlearnVERBfromADPaDETmodelNOUNsafelyADV,PUNCTwithoutADPrealADJharmNOUN
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1YouyouPRONCase=Nom Person=2 PronType=Prs3 learnnsubj
2cancanAUXVerbForm=Fin3 learnaux
3learnlearnVERBVerbForm=Infrootroot
4fromfromADP_6 modelcase
5aaDETDefinite=Ind PronType=Art6 modeldet
6modelmodelNOUNNumber=Sing3 learnobl
7safelysafelyADV_3 learnadvmod
8,,PUNCT_11 harmpunct
9withoutwithoutADP_11 harmcase
10realrealADJDegree=Pos11 harmamod
11harmharmNOUNNumber=Sing3 learnobl
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1learncanYoufrom a model; without real harm; safely propertyvn learn-14-2-1 «NP V PP.source»: agent=Agent from model=Source

Option 3

Simulations are useless because they're small

nsubjcopmarknsubjcopadvclrootSimulationsNOUNareAUXuselessADJbecauseSCONJtheyPRON'reAUXsmallADJ
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1SimulationssimulationNOUNNumber=Plur3 uselessnsubj
2arebeAUXMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin3 uselesscop
3uselessuselessADJDegree=Posrootroot
4becausebecauseSCONJ_7 smallmark
5theytheyPRONCase=Nom Number=Plur Person=3 PronType=Prs7 smallnsubj
6'rebeAUXMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin7 smallcop
7smallsmallADJDegree=Pos3 uselessadvcl
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1ispresentSimulationsuseless propertytree rules
2ispresentSimulations = simulationsmall propertytree rules

Option 4

Toy trains and real trains are the same

amodnsubjccamodconjcopdetrootToyADJtrainsNOUNandCCONJrealADJtrainsNOUNareAUXtheDETsameADJ
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1ToytoyADJDegree=Pos2 trainsamod
2trainstrainNOUNNumber=Plur8 samensubj
3andandCCONJ_5 trainscc
4realrealADJDegree=Pos5 trainsamod
5trainstrainNOUNNumber=Plur2 trainsconj
6arebeAUXMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin8 samecop
7thetheDETDefinite=Def PronType=Art8 samedet
8samesameADJDegree=Posrootroot
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1ispresentToy trains and real trainssame propertytree 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/26-who-you-are-mova/t5/