Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
Libraries:
Datasets
pandas
License:
premise
stringlengths
28
113
hypothesis
stringlengths
28
113
gold_label_log
class label
3 classes
gold_label_prag
class label
2 classes
spec_relation
stringclasses
8 values
item_type
stringclasses
2 values
trigger
stringclasses
1 value
lexemes
stringclasses
1 value
These computers or dresses would irritate Veronica.
These computers and dresses wouldn't both irritate Veronica.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
These computers and dresses wouldn't both irritate Veronica.
These computers or dresses would irritate Veronica.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
These computers or dresses would irritate Veronica.
These computers and dresses would irritate Veronica.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
These computers and dresses would irritate Veronica.
These computers or dresses would irritate Veronica.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
These computers and dresses wouldn't both irritate Veronica.
Neither these computers nor dresses would irritate Veronica.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither these computers nor dresses would irritate Veronica.
These computers and dresses wouldn't both irritate Veronica.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
These computers and dresses would irritate Veronica.
Neither these computers nor dresses would irritate Veronica.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither these computers nor dresses would irritate Veronica.
These computers and dresses would irritate Veronica.
2contradiction
2contradiction
opposite
control
connective
or - and
These computers or dresses would irritate Veronica.
Neither these computers nor dresses would irritate Veronica.
2contradiction
2contradiction
negation
control
connective
or - and
Neither these computers nor dresses would irritate Veronica.
These computers or dresses would irritate Veronica.
2contradiction
2contradiction
negation
control
connective
or - and
These computers and dresses would irritate Veronica.
These computers and dresses wouldn't both irritate Veronica.
2contradiction
2contradiction
negation
control
connective
or - and
These computers and dresses wouldn't both irritate Veronica.
These computers and dresses would irritate Veronica.
2contradiction
2contradiction
negation
control
connective
or - and
Those children or adults pressure Randolf to leave Monet.
Those children and adults don't both pressure Randolf to leave Monet.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
Those children and adults don't both pressure Randolf to leave Monet.
Those children or adults pressure Randolf to leave Monet.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
Those children or adults pressure Randolf to leave Monet.
Those children and adults pressure Randolf to leave Monet.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
Those children and adults pressure Randolf to leave Monet.
Those children or adults pressure Randolf to leave Monet.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
Those children and adults don't both pressure Randolf to leave Monet.
Neither those children nor adults pressure Randolf to leave Monet.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither those children nor adults pressure Randolf to leave Monet.
Those children and adults don't both pressure Randolf to leave Monet.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
Those children and adults pressure Randolf to leave Monet.
Neither those children nor adults pressure Randolf to leave Monet.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither those children nor adults pressure Randolf to leave Monet.
Those children and adults pressure Randolf to leave Monet.
2contradiction
2contradiction
opposite
control
connective
or - and
Those children or adults pressure Randolf to leave Monet.
Neither those children nor adults pressure Randolf to leave Monet.
2contradiction
2contradiction
negation
control
connective
or - and
Neither those children nor adults pressure Randolf to leave Monet.
Those children or adults pressure Randolf to leave Monet.
2contradiction
2contradiction
negation
control
connective
or - and
Those children and adults pressure Randolf to leave Monet.
Those children and adults don't both pressure Randolf to leave Monet.
2contradiction
2contradiction
negation
control
connective
or - and
Those children and adults don't both pressure Randolf to leave Monet.
Those children and adults pressure Randolf to leave Monet.
2contradiction
2contradiction
negation
control
connective
or - and
The Borgias or those adults had escaped from Ronald.
The Borgias and those adults hadn't both escaped from Ronald.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
The Borgias and those adults hadn't both escaped from Ronald.
The Borgias or those adults had escaped from Ronald.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
The Borgias or those adults had escaped from Ronald.
The Borgias and those adults had escaped from Ronald.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
The Borgias and those adults had escaped from Ronald.
The Borgias or those adults had escaped from Ronald.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
The Borgias and those adults hadn't both escaped from Ronald.
Neither the Borgias nor those adults had escaped from Ronald.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither the Borgias nor those adults had escaped from Ronald.
The Borgias and those adults hadn't both escaped from Ronald.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
The Borgias and those adults had escaped from Ronald.
Neither the Borgias nor those adults had escaped from Ronald.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither the Borgias nor those adults had escaped from Ronald.
The Borgias and those adults had escaped from Ronald.
2contradiction
2contradiction
opposite
control
connective
or - and
The Borgias or those adults had escaped from Ronald.
Neither the Borgias nor those adults had escaped from Ronald.
2contradiction
2contradiction
negation
control
connective
or - and
Neither the Borgias nor those adults had escaped from Ronald.
The Borgias or those adults had escaped from Ronald.
2contradiction
2contradiction
negation
control
connective
or - and
The Borgias and those adults had escaped from Ronald.
The Borgias and those adults hadn't both escaped from Ronald.
2contradiction
2contradiction
negation
control
connective
or - and
The Borgias and those adults hadn't both escaped from Ronald.
The Borgias and those adults had escaped from Ronald.
2contradiction
2contradiction
negation
control
connective
or - and
Those girls or senators fixed some skateboards.
Those girls and senators didn't both fix some skateboards.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
Those girls and senators didn't both fix some skateboards.
Those girls or senators fixed some skateboards.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
Those girls or senators fixed some skateboards.
Those girls and senators fixed some skateboards.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
Those girls and senators fixed some skateboards.
Those girls or senators fixed some skateboards.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
Those girls and senators didn't both fix some skateboards.
Neither those girls nor senators fixed some skateboards.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither those girls nor senators fixed some skateboards.
Those girls and senators didn't both fix some skateboards.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
Those girls and senators fixed some skateboards.
Neither those girls nor senators fixed some skateboards.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither those girls nor senators fixed some skateboards.
Those girls and senators fixed some skateboards.
2contradiction
2contradiction
opposite
control
connective
or - and
Those girls or senators fixed some skateboards.
Neither those girls nor senators fixed some skateboards.
2contradiction
2contradiction
negation
control
connective
or - and
Neither those girls nor senators fixed some skateboards.
Those girls or senators fixed some skateboards.
2contradiction
2contradiction
negation
control
connective
or - and
Those girls and senators fixed some skateboards.
Those girls and senators didn't both fix some skateboards.
2contradiction
2contradiction
negation
control
connective
or - and
Those girls and senators didn't both fix some skateboards.
Those girls and senators fixed some skateboards.
2contradiction
2contradiction
negation
control
connective
or - and
Women or those drivers had sung.
Women and those drivers hadn't both sung.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
Women and those drivers hadn't both sung.
Women or those drivers had sung.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
Women or those drivers had sung.
Women and those drivers had sung.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
Women and those drivers had sung.
Women or those drivers had sung.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
Women and those drivers hadn't both sung.
Neither women nor those drivers had sung.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither women nor those drivers had sung.
Women and those drivers hadn't both sung.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
Women and those drivers had sung.
Neither women nor those drivers had sung.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither women nor those drivers had sung.
Women and those drivers had sung.
2contradiction
2contradiction
opposite
control
connective
or - and
Women or those drivers had sung.
Neither women nor those drivers had sung.
2contradiction
2contradiction
negation
control
connective
or - and
Neither women nor those drivers had sung.
Women or those drivers had sung.
2contradiction
2contradiction
negation
control
connective
or - and
Women and those drivers had sung.
Women and those drivers hadn't both sung.
2contradiction
2contradiction
negation
control
connective
or - and
Women and those drivers hadn't both sung.
Women and those drivers had sung.
2contradiction
2contradiction
negation
control
connective
or - and
These mushrooms or these restaurants have stunned Tina.
These mushrooms and these restaurants haven't both stunned Tina.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
These mushrooms and these restaurants haven't both stunned Tina.
These mushrooms or these restaurants have stunned Tina.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
These mushrooms or these restaurants have stunned Tina.
These mushrooms and these restaurants have stunned Tina.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
These mushrooms and these restaurants have stunned Tina.
These mushrooms or these restaurants have stunned Tina.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
These mushrooms and these restaurants haven't both stunned Tina.
Neither these mushrooms nor these restaurants have stunned Tina.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither these mushrooms nor these restaurants have stunned Tina.
These mushrooms and these restaurants haven't both stunned Tina.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
These mushrooms and these restaurants have stunned Tina.
Neither these mushrooms nor these restaurants have stunned Tina.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither these mushrooms nor these restaurants have stunned Tina.
These mushrooms and these restaurants have stunned Tina.
2contradiction
2contradiction
opposite
control
connective
or - and
These mushrooms or these restaurants have stunned Tina.
Neither these mushrooms nor these restaurants have stunned Tina.
2contradiction
2contradiction
negation
control
connective
or - and
Neither these mushrooms nor these restaurants have stunned Tina.
These mushrooms or these restaurants have stunned Tina.
2contradiction
2contradiction
negation
control
connective
or - and
These mushrooms and these restaurants have stunned Tina.
These mushrooms and these restaurants haven't both stunned Tina.
2contradiction
2contradiction
negation
control
connective
or - and
These mushrooms and these restaurants haven't both stunned Tina.
These mushrooms and these restaurants have stunned Tina.
2contradiction
2contradiction
negation
control
connective
or - and
Those patients or the waiters will slump over.
Those patients and the waiters won't both slump over.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
Those patients and the waiters won't both slump over.
Those patients or the waiters will slump over.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
Those patients or the waiters will slump over.
Those patients and the waiters will slump over.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
Those patients and the waiters will slump over.
Those patients or the waiters will slump over.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
Those patients and the waiters won't both slump over.
Neither those patients nor the waiters will slump over.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither those patients nor the waiters will slump over.
Those patients and the waiters won't both slump over.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
Those patients and the waiters will slump over.
Neither those patients nor the waiters will slump over.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither those patients nor the waiters will slump over.
Those patients and the waiters will slump over.
2contradiction
2contradiction
opposite
control
connective
or - and
Those patients or the waiters will slump over.
Neither those patients nor the waiters will slump over.
2contradiction
2contradiction
negation
control
connective
or - and
Neither those patients nor the waiters will slump over.
Those patients or the waiters will slump over.
2contradiction
2contradiction
negation
control
connective
or - and
Those patients and the waiters will slump over.
Those patients and the waiters won't both slump over.
2contradiction
2contradiction
negation
control
connective
or - and
Those patients and the waiters won't both slump over.
Those patients and the waiters will slump over.
2contradiction
2contradiction
negation
control
connective
or - and
Those windows or these closets have shut.
Those windows and these closets haven't both shut.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
Those windows and these closets haven't both shut.
Those windows or these closets have shut.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
Those windows or these closets have shut.
Those windows and these closets have shut.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
Those windows and these closets have shut.
Those windows or these closets have shut.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and
Those windows and these closets haven't both shut.
Neither those windows nor these closets have shut.
1neutral
2contradiction
negated implicature_N
target
connective
or - and
Neither those windows nor these closets have shut.
Those windows and these closets haven't both shut.
0entailment
2contradiction
reverse negated implicature_N
target
connective
or - and
Those windows and these closets have shut.
Neither those windows nor these closets have shut.
2contradiction
2contradiction
opposite
control
connective
or - and
Neither those windows nor these closets have shut.
Those windows and these closets have shut.
2contradiction
2contradiction
opposite
control
connective
or - and
Those windows or these closets have shut.
Neither those windows nor these closets have shut.
2contradiction
2contradiction
negation
control
connective
or - and
Neither those windows nor these closets have shut.
Those windows or these closets have shut.
2contradiction
2contradiction
negation
control
connective
or - and
Those windows and these closets have shut.
Those windows and these closets haven't both shut.
2contradiction
2contradiction
negation
control
connective
or - and
Those windows and these closets haven't both shut.
Those windows and these closets have shut.
2contradiction
2contradiction
negation
control
connective
or - and
The adults or these women were hiding away.
The adults and these women weren't both hiding away.
1neutral
0entailment
implicature_PtoN
target
connective
or - and
The adults and these women weren't both hiding away.
The adults or these women were hiding away.
1neutral
0entailment
implicature_NtoP
target
connective
or - and
The adults or these women were hiding away.
The adults and these women were hiding away.
1neutral
2contradiction
negated implicature_P
target
connective
or - and
The adults and these women were hiding away.
The adults or these women were hiding away.
0entailment
2contradiction
reverse negated implicature_P
target
connective
or - and

Dataset Card for IMPPRES

Dataset Summary

Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.

Supported Tasks and Leaderboards

Natural Language Inference.

Languages

English.

Dataset Structure

Data Instances

The data consists of 2 configurations: implicature and presupposition. Each configuration consists of several different sub-datasets:

Pressupposition

  • all_n_presupposition
  • change_of_state
  • cleft_uniqueness
  • possessed_definites_existence
  • question_presupposition
  • both_presupposition
  • cleft_existence
  • only_presupposition
  • possessed_definites_uniqueness

Implicature

  • connectives
  • gradable_adjective
  • gradable_verb
  • modals
  • numerals_10_100
  • numerals_2_3
  • quantifiers

Each sentence type in IMPPRES is generated according to a template that specifies the linear order of the constituents in the sentence. The constituents are sampled from a vocabulary of over 3000 lexical items annotated with grammatical features needed to ensure wellformedness. We semiautomatically generate IMPPRES using a codebase developed by Warstadt et al. (2019a) and significantly expanded for the BLiMP dataset (Warstadt et al., 2019b).

Here is an instance of the raw presupposition data from any sub-dataset:

{
"sentence1": "All ten guys that proved to boast might have been divorcing.", 
"sentence2": "There are exactly ten guys that proved to boast.", 
"trigger": "modal", 
"presupposition": "positive", 
"gold_label": "entailment", 
"UID": "all_n_presupposition", 
"pairID": "9e", 
"paradigmID": 0
}

and the raw implicature data from any sub-dataset:

{
"sentence1": "That teenager couldn't yell.", 
"sentence2": "That teenager could yell.", 
"gold_label_log": "contradiction", 
"gold_label_prag": "contradiction", 
"spec_relation": "negation", 
"item_type": "control", 
"trigger": "modal", 
"lexemes": "can - have to"
}

Data Fields

Presupposition

There is a slight mapping from the raw data fields in the presupposition sub-datasets and the fields appearing in the HuggingFace Datasets. When dealing with the HF Dataset, the following mapping of fields happens:

"premise" -> "sentence1"
"hypothesis"-> "sentence2"
"trigger" -> "trigger" or "Not_In_Example"
"trigger1" -> "trigger1" or "Not_In_Example"
"trigger2" -> "trigger2" or "Not_In_Example"
"presupposition" -> "presupposition" or "Not_In_Example"
"gold_label" -> "gold_label"
"UID" -> "UID"
"pairID" -> "pairID"
"paradigmID" -> "paradigmID"

For the most part, the majority of the raw fields remain unchanged. However, when it comes to the various trigger fields, a new mapping was introduced. There are some examples in the dataset that only have the trigger field while other examples have the trigger1 and trigger2 field without the trigger or presupposition field. Nominally, most examples look like the example in the Data Instances section above. Occassionally, however, some examples will look like:

{
'sentence1': 'Did that committee know when Lissa walked through the cafe?', 
'sentence2': 'That committee knew when Lissa walked through the cafe.', 
'trigger1': 'interrogative', 
'trigger2': 'unembedded', 
'gold_label': 'neutral', 
'control_item': True, 
'UID': 'question_presupposition', 
'pairID': '1821n', 
'paradigmID': 95
}

In this example, trigger1 and trigger2 appear and presupposition and trigger are removed. This maintains the length of the dictionary. To account for these examples, we have thus introduced the mapping above such that all examples accessed through the HF Datasets interface will have the same size as well as the same fields. In the event that an example does not have a value for one of the fields, the field is maintained in the dictionary but given a value of Not_In_Example.

To illustrate this point, the example given in the Data Instances section above would look like the following in the HF Datasets:

{
"premise": "All ten guys that proved to boast might have been divorcing.", 
"hypothesis": "There are exactly ten guys that proved to boast.", 
"trigger": "modal",
"trigger1":  "Not_In_Example",
"trigger2": "Not_In_Example"
"presupposition": "positive", 
"gold_label": "entailment", 
"UID": "all_n_presupposition", 
"pairID": "9e", 
"paradigmID": 0
}

Below is description of the fields:

"premise": The premise. 
"hypothesis": The hypothesis. 
"trigger": A detailed discussion of trigger types appears in the paper.
"trigger1":  A detailed discussion of trigger types appears in the paper.
"trigger2": A detailed discussion of trigger types appears in the paper.
"presupposition": positive or negative. 
"gold_label": Corresponds to entailment, contradiction, or neutral. 
"UID": Unique id. 
"pairID": Sentence pair ID.
"paradigmID": ?

It is not immediately clear what the difference is between trigger, trigger1, and trigger2 is or what the paradigmID refers to.

Implicature

The implicature fields only have the mapping below:

"premise" -> "sentence1"
"hypothesis"-> "sentence2"

Here is a description of the fields:

"premise": The premise. 
"hypothesis": The hypothesis. 
"gold_label_log": Gold label for a logical reading of the sentence pair.
"gold_label_prag": Gold label for a pragmatic reading of the sentence pair.
"spec_relation": ?
"item_type": ?
"trigger": A detailed discussion of trigger types appears in the paper.
"lexemes": ? 

Data Splits

As the dataset was created to test already trained models, the only split that exists is for testing.

Dataset Creation

Curation Rationale

IMPPRES was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

The annotations were generated semi-automatically.

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

[More Information Needed]

Licensing Information

IMPPRES is available under a Creative Commons Attribution-NonCommercial 4.0 International Public License ("The License"). You may not use these files except in compliance with the License. Please see the LICENSE file for more information before you use the dataset.

Citation Information

@inproceedings{jeretic-etal-2020-natural,
    title = "Are Natural Language Inference Models {IMPPRESsive}? {L}earning {IMPlicature} and {PRESupposition}",
    author = "Jereti\v{c}, Paloma  and
      Warstadt, Alex  and
      Bhooshan, Suvrat  and
      Williams, Adina",
    booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
    month = jul,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.acl-main.768",
    doi = "10.18653/v1/2020.acl-main.768",
    pages = "8690--8705",
    abstract = "Natural language inference (NLI) is an increasingly important task for natural language understanding, which requires one to infer whether a sentence entails another. However, the ability of NLI models to make pragmatic inferences remains understudied. We create an IMPlicature and PRESupposition diagnostic dataset (IMPPRES), consisting of 32K semi-automatically generated sentence pairs illustrating well-studied pragmatic inference types. We use IMPPRES to evaluate whether BERT, InferSent, and BOW NLI models trained on MultiNLI (Williams et al., 2018) learn to make pragmatic inferences. Although MultiNLI appears to contain very few pairs illustrating these inference types, we find that BERT learns to draw pragmatic inferences. It reliably treats scalar implicatures triggered by {``}some{''} as entailments. For some presupposition triggers like {``}only{''}, BERT reliably recognizes the presupposition as an entailment, even when the trigger is embedded under an entailment canceling operator like negation. BOW and InferSent show weaker evidence of pragmatic reasoning. We conclude that NLI training encourages models to learn some, but not all, pragmatic inferences.",
}

Contributions

Thanks to @aclifton314 for adding this dataset.

Downloads last month
373

Models trained or fine-tuned on facebook/imppres