verb/index.html at master · tdavislab/verb · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <title>Visualizing Word Vector Biases</title>
  <link rel="stylesheet" href="https://stackpath.bootstrapcdn.com/bootstrap/4.5.0/css/bootstrap.min.css">
  <link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.3.0/css/font-awesome.min.css">
  <link rel="stylesheet" href="../static/css/style.css">
  <link rel="icon" href="data:;base64,iVBORw0KGgo=">
</head>
<body>
<div class="container">
  <div class="jumbotron">
    <h1 class="text-center">A Visual Tour of Bias Mitigation Techniques for Word Representations</h1>
    <h6 class="text-center">Archit Rathore, Sunipa Dev, Jeff M. Phillips, Vivek Srikumar, Bei Wang</h6>
  </div>
  <div class="row"></div>
  <div class="p-4 text-center">
    Word vector embeddings have been shown to contain and amplify biases in data they are extracted from. Consequently, many techniques have been
    proposed to identify, mitigate, and attenuate these biases in word representations. In this tutorial, we review a collection of the
    state-of-the-art debiasing techniques. In particular, we provide an open source visualization tool and offer hands-on experience in exploring the
    effects of these debiasing techniques on the geometry of the high-dimensional word vectors. Via a web-based tool, we decompose each technique into
    interpretable sequences of operations, and study their effect on the word vectors using dimensionality reduction and interactive visual
    exploration.
  </div>
  <div class="row">
    <div class="p-4">
      <img src="static/assets/prototype.gif" class="img-fluid">
    </div>
  </div>

</div>
</body>
</html>