Changes between Version 24 and Version 25 of AnnotationManual/CZSNLI


Ignore:
Timestamp:
Mar 5, 2024, 1:55:07 PM (2 months ago)
Author:
Zuzana Nevěřilová
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • AnnotationManual/CZSNLI

    v24 v25  
    1515 * one or both sentences from the pair **do not make sense** (**bad translation**)
    1616
    17 The annotation is made via the [https://nlp.fi.muni.cz/projekty/labelstudio LabelStudio] tool. The procedure requires a double login process. 
     17The annotation is made via the [https://nlp.fi.muni.cz/projekty/labelstudio LabelStudio] tool. The procedure requires a double login process.
    1818
    1919First, log into the tool using the credentials obtained by email (`student@localhost`).
     
    2222
    2323=== How to annotate one sample ===
    24 
    2524!LabelStudio limitation: Do not simultaneously open one labeling task in two tabs or windows.
    2625
    2726The annotation steps are the following.
    2827
    29 ==== 1. Rule out bad translations
    30 
     28==== 1. Rule out bad translations ====
    3129Read the premise and the hypothesis and decide whether they make sense. If not, select **bad translation** and continue to another pair. You can check your understanding by revealing the original texts.
    3230
    33 ==== 2. Decide the sentence relationship
    34 
     31==== 2. Decide the sentence relationship ====
    3532If the sentences make sense, decide whether the relation between them is **entailment**, **contradiction**, or **neutral**.
    3633
     
    4037
    4138Typically:
    42 1. If the hypothesis is a more general statement than the premise, the result is **entailment**.
    43 1. If the premise adds new information or denotes an element of a hypothesis by a more specific name, it is **neutral** (cannot say).
    44 1. If the hypothesis and the premise describe a non-related action with different actors or they describe mutually exclusive actions, it is a **contradiction**.
     39
     40 1. If the hypothesis is a more general statement than the premise, the result is **entailment**.
     41 1. If the premise adds new information or denotes an element of a hypothesis by a more specific name, it is **neutral** (cannot say).
     42 1. If the hypothesis and the premise describe a non-related action with different actors or they describe mutually exclusive actions, it is a **contradiction**.
    4543
    4644==== 3. Select relevant parts ====
    47 
    4845[[Image(entailment1.png,width=50%,right)]]
    4946
     
    5451Mark one or more words (the tool selects whole words and removes trailing spaces from the selection).
    5552
    56 In case of **entailment**, mark the elements that are the same thing but are described with different words. 
     53In case of **entailment**, mark the elements that are the same thing but are described with different words.
    5754
    5855In case of **contradiction**, mark the elements that are mutually exclusive (e.g., man-woman) in the premise and hypothesis. If the hypothesis and the premise are completely unrelated, do not mark anything.
     
    6057In case of **neutral**, there is often some new information in the hypothesis, for instance, a more specific term (woman-human) or some independent term. In the neutral pair, the hypothesis may have one element with no relation to the premise.
    6158
     59You can remove selection by selecting it and clicking a red garbage bin in the Selection details tab.
     60
    6261==== 4. Create a supporting relation ====
     62Set up the relation between the elements in the premise and hypothesis:
    6363
    64 Set up the relation between the elements in the premise and hypothesis:
    65 1. Click the relevant element in premise (1).
    66 1. select the **Info** tab and click the **Create relation** button, a chain icon (2).
    67 1. Next, click the relevant part in the hypothesis (3).
     64 1. Click the relevant element in premise (1).
     65 1. select the **Info** tab and click the **Create relation** button, a chain icon (2).
     66 1. Next, click the relevant part in the hypothesis (3).
    6867
    6968[[Image(entailment2.png,width=50%,center)]]
    70 
    7169
    7270[[Image(entailment3.png,width=50%,center)]]
    7371
    7472Then select the relation type: click on the **Relations** tab, and select the relation by clicking the triple dot button. Next, click into the **Select labels** box and choose a relation type:
    75 * In the case of **entailment**, the relation would be green, i.e., generalization or similar.
    76 * In the case of **contradiction**, the relation would be red, i.e., exclusion.
    77 * In the case of **neutral**, the relation would be orange.
     73
     74 * In the case of **entailment**, the relation would be green, i.e., generalization or similar.
     75 * In the case of **contradiction**, the relation would be red, i.e., exclusion.
     76 * In the case of **neutral**, the relation would be orange.
    7877
    7978It does not matter if you mark prepositional phrases (e.g., "''v obýváku''") or noun phrases (e.g., "''obýváku''").
     79
     80You can remove a relation by clicking a red garbage bin in the Relations tab.
    8081
    8182[[Image(entailment4.png,width=15%,center)]]
    8283
    8384=== Examples of annotations ===
     85==== Example 1 ====
     86  **Premise:** '' Vousatý muž v malířské čepici hraje na xylofon na straně rušné lávky.''
    8487
    85 ==== Example 1 ====
     88  **Hypothesis:** '' Muž s plnovousem hraje na nástroj na chodníku.''
    8689
    87  **Premise:** ''Vousatý muž v malířské čepici hraje na xylofon na straně rušné lávky.''
    88 
    89  **Hypothesis:** ''Muž s plnovousem hraje na nástroj na chodníku.''
    90 
    91  The information in the hypothesis is more general, so mark it as **entailment**. For support of the decision, mark "xylofon" and "nástroj" and connect with the **generalization** relation, and "Vousatý muž" and "Muž s plnovousem" with the **similar** relation.
     90  The information in the hypothesis is more general, so mark it as **entailment**. For support of the decision, mark "xylofon" and "nástroj" and connect with the **generalization** relation, and "Vousatý muž" and "Muž s plnovousem" with the **similar** relation.
    9291
    9392==== Example 2 ====
     93  **Premise:** '' Chlapec je zajištěn mezi dvěma bungee lany a visí ve vzduchu.''
    9494
    95  **Premise:** ''Chlapec je zajištěn mezi dvěma bungee lany a visí ve vzduchu.''
     95  **Hypothesis:** '' Chlapec se prochází v parku.''
    9696
    97  **Hypothesis:** ''Chlapec se prochází v parku.''
    98 
    99  The boy cannot do both things. Mark as **contradiction**. To support the decision, mark "visí" (is suspended) and "se prochází" (is walking) as **exclusion**.
     97  The boy cannot do both things. Mark as **contradiction**. To support the decision, mark "visí" (is suspended) and "se prochází" (is walking) as **exclusion**.
    10098
    10199==== Example 3 ====
     100  **Premise:** '' Surfař v zeleném pruhovaném neoprenu jede na vlně.''
    102101
    103  **Premise:** ''Surfař v zeleném pruhovaném neoprenu jede na vlně.''
     102  **Hypothesis:** '' Muž, který surfuje, má zelený pruhovaný neopren se svým psem.''
    104103
    105  **Hypothesis:** ''Muž, který surfuje, má zelený pruhovaný neopren se svým psem.''
    106 
    107  The information about the dog is more specific, therefore mark as **neutral**. To support the decision, mark "se svým psem" without relation to the premise.
     104  The information about the dog is more specific, therefore mark as **neutral**. To support the decision, mark "se svým psem" without relation to the premise.
    108105
    109106==== Example 4 ====
     107  **Premise:** '' Dvě dámy se střetnou se třemi osly.''
    110108
    111  **Premise:** ''Dvě dámy se střetnou se třemi osly.''
     109  **Hypothesis:** '' Dvě dámy půjdou nakrmit tři osly do zoo.''
    112110
    113  **Hypothesis:** ''Dvě dámy půjdou nakrmit tři osly do zoo.''
     111  We cannot tell whether the ladies will feed the donkeys if we only know they meet the donkeys.
    114112
    115  We cannot tell whether the ladies will feed the donkeys if we only know they meet the donkeys.
    116 
    117  Mark as **neutral**. For support of the decision, mark "se střetnou" (meet) and "půjdou nakrmit" (will feed) as **independence**.
     113  Mark as **neutral**. For support of the decision, mark "se střetnou" (meet) and "půjdou nakrmit" (will feed) as **independence**.
    118114
    119115==== Example 5 ====
     116  **Premise:** '' Starší muž hrající na podivný nástroj podobný kytaře na něčem, co vypadá jako park nebo otevřené prostranství.''
    120117
    121  **Premise:** ''Starší muž hrající na podivný nástroj podobný kytaře na něčem, co vypadá jako park nebo otevřené prostranství.''
     118  **Hypothesis:** '' Žena hraje v obýváku na klavír.''
    122119
    123  **Hypothesis:** ''Žena hraje v obýváku na klavír.''
    124 
    125  Since a man is not a woman, mark it as **contradiction**. To support the decision, mark "Starší muž" and "Žena" with the **exclusion** relation. Mark also "park nebo otevřené prostranství" and "obýváku" with the **exclusion** relation.
     120  Since a man is not a woman, mark it as **contradiction**. To support the decision, mark "Starší muž" and "Žena" with the **exclusion** relation. Mark also "park nebo otevřené prostranství" and "obýváku" with the **exclusion** relation.
    126121
    127122==== Example 6 ====
     123  **Premise:** '' Kůň vlevo má bílou hřívu.''
    128124
    129  **Premise:** ''Kůň vlevo má bílou hřívu.''
     125  **Hypothesis:** '' Je tam víc než jeden kůň.''
    130126
    131  **Hypothesis:** ''Je tam víc než jeden kůň.''
    132 
    133  From the premise, it follows that there are more horses. Mark the relation as **entailment** and connect "Kůň vlevo" (the left horse) and "víc než jeden kůň" (more than one horse) with the **generalization** relation.
    134 
     127  From the premise, it follows that there are more horses. Mark the relation as **entailment** and connect "Kůň vlevo" (the left horse) and "víc než jeden kůň" (more than one horse) with the **generalization** relation.
    135128
    136129==== Example 7 ====
     130  **Premise:** '' Několik lidí pohybujících nějakou stavbou.''
    137131
    138  **Premise:** ''Několik lidí pohybujících nějakou stavbou.''
     132  **Hypothesis:** '' Několik lidí stěhuje kůlnu na nářadí do zadní části dvora.''
    139133
    140  **Hypothesis:** ''Několik lidí stěhuje kůlnu na nářadí do zadní části dvora.''
    141 
    142  We don't know the nature of the structure. So, mark it as **neutral**. To support the decision, mark "nějakou stavbou" and "kůlnu na nářadí" with the **specification** relation.
     134  We don't know the nature of the structure. So, mark it as **neutral**. To support the decision, mark "nějakou stavbou" and "kůlnu na nářadí" with the **specification** relation.
    143135
    144136==== Example 8 ====
     137  **Premise:** '' Ta mladá žena s šátkem dává někomu dárek.''
    145138
    146  **Premise:** ''Ta mladá žena s šátkem dává někomu dárek.''
     139  **Hypothesis:** '' Současnost je malá.''
    147140
    148  **Hypothesis:** ''Současnost je malá.''
     141  Apparently, the two sentences have no relation; moreover, the hypothesis is suspicious.
    149142
    150  Apparently, the two sentences have no relation; moreover, the hypothesis is suspicious.
    151 
    152  Check the English origin (The young woman with the scarf is giving a present to someone. The present is small.) and mark it as **bad translation**.
     143  Check the English origin (The young woman with the scarf is giving a present to someone. The present is small.) and mark it as **bad translation**.
    153144
    154145==== Example 9 ====
     146  **Premise:** '' Žena a dvě malé holčičky slaví narozeniny s mužem přes webovou kameru.''
    155147
    156  **Premise:** ''Žena a dvě malé holčičky slaví narozeniny s mužem přes webovou kameru.''
     148  **Hypothesis:** '' Muž běží po ulici''
    157149
    158  **Hypothesis:** ''Muž běží po ulici''
    159 
    160  Completely unrelated stories. Annotated as **contradiction** and do not mark any relations.
     150  Completely unrelated stories. Annotated as **contradiction** and do not mark any relations.
    161151
    162152=== Submit ===
    163 
    164153Finally, **Submit** your annotation.
    165154
     
    167156
    168157=== Further information ===
    169 
    170158You can check the instructions by using the **Instructions** button on the top when you need it.
    171159
     
    173161
    174162You can try a short [https://nlp.fi.muni.cz/trac/research/wiki/AnnotationManual/CZSNLI/training training] on your own.
    175