Skip to content
This repository has been archived by the owner on Jan 20, 2023. It is now read-only.

Add blurb about bias to MLM demo. #564

Merged
merged 4 commits into from Oct 16, 2020
Merged

Add blurb about bias to MLM demo. #564

merged 4 commits into from Oct 16, 2020

Conversation

schmmd
Copy link
Member

@schmmd schmmd commented Oct 15, 2020

No description provided.

@schmmd schmmd changed the title Add blurb about bias Add blurb about bias to MLM demo. Oct 15, 2020
@@ -154,7 +154,9 @@ const description = (
<span>
<p>Masked language modeling is a fill-in-the-blank task, where a model uses the context words surrounding a [MASK] token to try to predict what the [MASK] word should be.</p>
<p>The model shown here is <a href="https://arxiv.org/abs/1810.04805" target="_blank" rel="noopener noreferrer">BERT</a>, the first large transformer to be trained on this task.
Enter text with one or more "[MASK]" tokens and the model will generate the most likely substitution for each.</p>
Enter text with one or more "[MASK]" tokens and the model will generate the most likely substitution for each. Note that the default example demonstrates how BERT demonstrates gender bias, in that it thinks the doctor is more likely a man ("his")
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggestion: Note that in the default example, BERT demonstrates gender bias...

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, that's much better. I half expected an improvement from you here...

@epwalsh epwalsh merged commit 8db4a1a into master Oct 16, 2020
@epwalsh epwalsh deleted the add-blurb-about-bias branch October 16, 2020 16:18
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants