| 1 |
Bislama: An Introduction to the National Language of Vanuatu
|
|
|
|
BASE
|
|
Show details
|
|
|
| 2 |
兒童華語教材中的飲食文化分析:以《學華語向前走》為例 ; Analysis of Food Culture of Chinese Teaching Material for Children — With Reference to “Let’s Learn Chinese”
|
|
|
|
BASE
|
|
Show details
|
|
|
| 3 |
Sentence Level Embedding Detoxification via Toxic Component Removal ...
|
|
: University of Virginia, 2022
|
|
Abstract:
When state of the art pre-trained language models are trained on internet discourse, they may unintentionally encode and perpetuate some of the hateful and toxic behavior known to exist within the training data. We hypothesize that the encoding of toxicity can be explained by a "toxic subspace" within sentence-level language model embeddings, the existence of which suggests that toxic features follow some underlying pattern and are thus removable. To identify this toxic subspace, we propose a method to find patterns in toxic lexical markers. Through our experiments, we demonstrate that the subspace we find can separate toxic and non-toxic examples in the latent space. We use this subspace to extract a toxic component within sentence embeddings, and demonstrate that the removal of this component causes the number of toxic representations to decrease. These results are promising and suggest that encoders can be reasonably detoxified. ...
|
|
Keyword:
Language Models; NLP for Social Good; Representation Learning
|
|
URL: https://dx.doi.org/10.18130/5pch-v272
https://libraetd.lib.virginia.edu/public_view/k06988420
|
|
BASE
|
|
Hide details
|
|
|
| 4 |
It's a Two-way Street: Informing Irish Pre-sessional EAP Programs with a Needs Analysis of Irish Higher Education
|
|
|
|
BASE
|
|
Show details
|
|
|
|
|
