--- title: Many chatbots are trained to be helpful and pleasant, and sometimes… — Hallucinations: confident made-up answers · tree and pragmatics description: Many chatbots are trained to be helpful and pleasant, and sometimes that tips over into agreeing, praising and inventing. Who acts: Many chatbots, that. A sentence of the lesson «Hallucinations: confident made-up answers» of the ai course of Mova's primary school with its dependency tree and the pra --- [Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/corpora/school/ai/5-hallucinations-confident-made-up-answers/2/) [home](https://md.mova.elamur.ai/free/en/) / [corpora](https://md.mova.elamur.ai/free/en/corpora/) / [school](https://md.mova.elamur.ai/free/en/corpora/school/) / [AI](https://md.mova.elamur.ai/free/en/corpora/school/ai/) / [Hallucinations: confident made-up answ…](https://md.mova.elamur.ai/free/en/corpora/school/ai/5-hallucinations-confident-made-up-answers/) / 2 # Hallucinations: confident made-up answers, sentence 2 Many chatbots are trained to be helpful and pleasant, and sometimes that tips over into agreeing, praising and inventing. **Who acts:** Many chatbots, that · **What happens:** train, be, tip over, agree, praise, invent · **To what, where:** Many chatbots This is a sentence of the lesson «Hallucinations: confident made-up answers» of the ai course of Mova's primary school. The page gives its syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause: who acts (Many chatbots, that), what happens, when and where, as Mova's frames read it from the tree. ## The tree {#tree} ```conllu # sent_id = school:ai:005-hallucinations-confident-made-up-answers.lesson:2 # text = Many chatbots are trained to be helpful and pleasant, and sometimes that tips over into agreeing, praising and inventing. # source = school lesson (ours) # course = ai # lesson = 005-hallucinations-confident-made-up-answers.lesson # annotation = Mova annot: Rust (en) → Haiku 5.5 (medium, Batches API) patches; gates UD 2.18; agreed by treeagree rules (Stanza, Trankit, spaCy trf): R3-conj-first ×1, R6-punct-agree ×1, R7-rel-agree ×2 1 Many many ADJ JJ Degree=Pos 2 amod _ _ 2 chatbots chatbot NOUN NNS Number=Plur 4 nsubj:pass _ _ 3 are be AUX VBP Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin 4 aux:pass _ _ 4 trained train VERB VBN Tense=Past|VerbForm=Part|Voice=Pass 0 root _ _ 5 to to PART TO _ 7 mark _ _ 6 be be AUX VB VerbForm=Inf 7 cop _ _ 7 helpful helpful ADJ JJ Degree=Pos 4 xcomp _ _ 8 and and CCONJ CC _ 9 cc _ _ 9 pleasant pleasant ADJ JJ Degree=Pos 7 conj _ _ 10 , , PUNCT , _ 14 punct _ Agree=R6-punct-agree|Haiku=17:punct 11 and and CCONJ CC _ 14 cc _ _ 12 sometimes sometimes ADV RB PronType=Ind 14 advmod _ _ 13 that that PRON DT Number=Sing|PronType=Dem 14 nsubj _ _ 14 tips tip VERB VBZ Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin 4 conj _ _ 15 over over ADV RB _ 14 compound:prt _ _ 16 into into SCONJ IN _ 17 mark _ Agree=R7-rel-agree|Haiku=17:case|HaikuPos=ADP:IN:_ 17 agreeing agree VERB VBG Tense=Pres|VerbForm=Part 14 advcl _ Agree=R7-rel-agree|Haiku=14:obl 18 , , PUNCT , _ 19 punct _ _ 19 praising praise VERB VBG Tense=Pres|VerbForm=Part 17 conj _ _ 20 and and CCONJ CC _ 21 cc _ _ 21 inventing invent VERB VBG Tense=Pres|VerbForm=Part 17 conj _ Agree=R3-conj-first|Haiku=19:conj 22 . . PUNCT . _ 4 punct _ _ ``` ## Pragmatics {#pragmatics} - form: statement - act: assert - voice: narrator - polarity: none - teacher: kept the draft | # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 1 | **train** | present · passive | | Many chatbots | | | | | vn hire-13.5.3 «NP V NP»: subject=Theme | | 2 | **is** | to-infinitive | Many chatbots *not said here* | | | | | helpful *property* | tree rules | | 3 | **be** | | Many chatbots *not said here* | | | | | | tree rules | | 4 | **tip over** | present | that | | | | | sometimes *property* | vn throw-17.1-1-1 «NP V NP»: agent=Agent | | 5 | **agree** *speech: agree* | gerund | that *not said here* | | | | | | vn correspond-36.1-1 «NP V»: agent=Agent | | 6 | **praise** | gerund | that *not said here* | | | | | | vn judgment-33.1-1 «NP V NP»: agent=Agent | | 7 | **invent** | gerund | that *not said here* | | | | | | vn create-26.4-1 «NP V»: agent=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. **Licence: [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/)**, as the whole of Mova corpus 0.4, of which this treebank (the agreed school, `mova-school-agreed-en.conllu`) is part. The lessons were written for Mova's primary school. Sentence `school:ai:005-hallucinations-confident-made-up-answers.lesson:2`. **The tree:** Mova annot: Rust (en) → Haiku 5.5 (medium, Batches API) patches; gates UD 2.18; agreed by treeagree rules (Stanza, Trankit, spaCy trf): R3-conj-first ×1, R6-punct-agree ×1, R7-rel-agree ×2. **The frames** are read from the tree by Mova's frames reader (`prag`) when this page is built, the lesson read in order, so a pronoun may be resolved from an earlier sentence. See the [release 0.4](https://md.mova.elamur.ai/free/en/releases/0.4/). - ← previous: [Hallucinations: confident made-up answ…, 0](https://md.mova.elamur.ai/free/en/corpora/school/ai/5-hallucinations-confident-made-up-answers/0/) - next →: [Hallucinations: confident made-up answ…, 3](https://md.mova.elamur.ai/free/en/corpora/school/ai/5-hallucinations-confident-made-up-answers/3/) - ↑ back to [the lesson](https://md.mova.elamur.ai/free/en/corpora/school/ai/5-hallucinations-confident-made-up-answers/)