Text Classification
Transformers
PyTorch
English
deberta-v2
Inference Endpoints
ikrysinska commited on
Commit
0783058
1 Parent(s): 73ed339

Update README.md

Browse files

Add Jan's thoughts about model's misuse

Files changed (1) hide show
  1. README.md +13 -6
README.md CHANGED
@@ -14,17 +14,17 @@ pipeline_tag: text-classification
14
  <!-- Provide a longer summary of what this model is. -->
15
  This is a DeBERTa-v3-base-tasksource-nli model with an adapter trained on [More Information Needed], which contains X pairs of a tweet and a conspiracy theory along with class labels: support, deny, neutral. The model was finetuned for text classification to predict whether a tweet supports a given conspiracy theory or not. The model was trained on tweets related to six common COVID-19 conspiracy theories.
16
 
17
- 1. **Vaccines are unsafe.** The coronavirus vaccine is either unsafe or part of a larger plot to control people or reduce the population.
18
 
19
- 2. **Governments and politicians spread misinformation.** Politicians or government agencies are intentionally spreading false information, or they have some other motive for the way they are responding to the coronavirus.
20
 
21
- 3. **The Chinese intentionally spread the virus.** The Chinese government intentionally created or spread the coronavirus to harm other countries.
22
 
23
- 4. **Deliberate strategy to create economic instability or benefit large corporations.** The coronavirus or the government's response to it is a deliberate strategy to create economic instability or to benefit large corporations over small businesses.
24
 
25
- 5. **Public was intentionally misled about the true nature of the virus and prevention.** The public is being intentionally misled about the true nature of the Coronavirus, its risks, or the efficacy of certain treatments or prevention methods.
26
 
27
- 6. **Human made and bioweapon.** The Coronavirus was created intentionally, made by humans, or as a bioweapon.
28
 
29
 
30
  This model is suitable for English only.
@@ -54,12 +54,19 @@ This model is suitable for English only.
54
 
55
  ### Out-of-Scope Use
56
 
 
 
 
 
57
  <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
58
 
59
  [More Information Needed]
60
 
61
  ## Bias, Risks, and Limitations
62
 
 
 
 
63
  <!-- This section is meant to convey both technical and sociotechnical limitations. -->
64
 
65
  [More Information Needed]
 
14
  <!-- Provide a longer summary of what this model is. -->
15
  This is a DeBERTa-v3-base-tasksource-nli model with an adapter trained on [More Information Needed], which contains X pairs of a tweet and a conspiracy theory along with class labels: support, deny, neutral. The model was finetuned for text classification to predict whether a tweet supports a given conspiracy theory or not. The model was trained on tweets related to six common COVID-19 conspiracy theories.
16
 
17
+ 1. **CT6: Vaccines are unsafe.** The coronavirus vaccine is either unsafe or part of a larger plot to control people or reduce the population.
18
 
19
+ 2. **CT4: Governments and politicians spread misinformation.** Politicians or government agencies are intentionally spreading false information, or they have some other motive for the way they are responding to the coronavirus.
20
 
21
+ 3. **CT5: The Chinese intentionally spread the virus.** The Chinese government intentionally created or spread the coronavirus to harm other countries.
22
 
23
+ 4. **CT1: Deliberate strategy to create economic instability or benefit large corporations.** The coronavirus or the government's response to it is a deliberate strategy to create economic instability or to benefit large corporations over small businesses.
24
 
25
+ 5. **CT2: Public was intentionally misled about the true nature of the virus and prevention.** The public is being intentionally misled about the true nature of the Coronavirus, its risks, or the efficacy of certain treatments or prevention methods.
26
 
27
+ 6. **CT3: Human made and bioweapon.** The Coronavirus was created intentionally, made by humans, or as a bioweapon.
28
 
29
 
30
  This model is suitable for English only.
 
54
 
55
  ### Out-of-Scope Use
56
 
57
+ - spreading/generating tweets that support a given conspiracy theory
58
+ - amplifying echo chambers of social subnetworks susceptible to believe in conspiracy theories
59
+
60
+
61
  <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
62
 
63
  [More Information Needed]
64
 
65
  ## Bias, Risks, and Limitations
66
 
67
+ - results are distorted for the conspiracy theories out of the training dataset
68
+ - unintentional stifling of legitimate public discourse (elimination of discussion that resembles conspiracy theories from social subnetworks)
69
+ - bias: text style, economic status...
70
  <!-- This section is meant to convey both technical and sociotechnical limitations. -->
71
 
72
  [More Information Needed]