sileod commited on
Commit
08a87f7
·
1 Parent(s): fcfddd0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -31,6 +31,7 @@ tags:
31
  - logical reasoning
32
  - soft logic
33
  - nli
 
34
  - natural-language-inference
35
  - reasoning
36
  - logic
@@ -84,7 +85,7 @@ train-eval-index:
84
 
85
  # Dataset accompanying the "Probing neural language models for understanding of words of estimative probability" article
86
 
87
- This dataset tests the capabilities of language models to correctly capture the meaning of words denoting probabilities (WEP), e.g. words like "probably", "maybe", "surely", "impossible".
88
 
89
  We used probabilitic soft logic to combine probabilistic statements expressed with WEP (WEP-Reasoning) and we also used the UNLI dataset (https://nlp.jhu.edu/unli/) to directly check whether models can detect the WEP matching human-annotated probabilities according to [Fagen-Ulmschneider, 2018](https://github.com/wadefagen/datasets/tree/master/Perception-of-Probability-Words).
90
  The dataset can be used as natural language inference data (context, premise, label) or multiple choice question answering (context,valid_hypothesis, invalid_hypothesis).
 
31
  - logical reasoning
32
  - soft logic
33
  - nli
34
+ - verbal probabilities
35
  - natural-language-inference
36
  - reasoning
37
  - logic
 
85
 
86
  # Dataset accompanying the "Probing neural language models for understanding of words of estimative probability" article
87
 
88
+ This dataset tests the capabilities of language models to correctly capture the meaning of words denoting probabilities (WEP, also called verbal probabilities), e.g. words like "probably", "maybe", "surely", "impossible".
89
 
90
  We used probabilitic soft logic to combine probabilistic statements expressed with WEP (WEP-Reasoning) and we also used the UNLI dataset (https://nlp.jhu.edu/unli/) to directly check whether models can detect the WEP matching human-annotated probabilities according to [Fagen-Ulmschneider, 2018](https://github.com/wadefagen/datasets/tree/master/Perception-of-Probability-Words).
91
  The dataset can be used as natural language inference data (context, premise, label) or multiple choice question answering (context,valid_hypothesis, invalid_hypothesis).