Skip to content

Commit

Permalink
Update index.md
Browse files Browse the repository at this point in the history
  • Loading branch information
andyz245 authored Feb 2, 2024
1 parent 05b492a commit 79edda4
Showing 1 changed file with 5 additions and 7 deletions.
12 changes: 5 additions & 7 deletions index.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,16 +4,14 @@

# Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks

<div style="display: flex; justify-content: center; align-items: center;">
<div style="flex: 1; text-align: center; margin: 0 10px;">
<div style="display: flex; justify-content: center; align-items: flex-start;">
<div style="margin-right: 10px;">
<img src="figures/gpt4.png" alt="GPT-4" width="300"/>
<br>
<em>Fig.1 GPT-4 safety filters can be bypassed by jailbreaks!</em>
<p style="text-align: center;"><em>Fig.1 GPT-4 safety filters can be bypassed by jailbreaks!</em></p>
</div>
<div style="flex: 1; text-align: center; margin: 0 10px;">
<div>
<img src="figures/gpt4_rpo.png" alt="RPO" width="300"/>
<br>
<em>Fig.2 RPO enforces harmless responses even after jailbreaks</em>
<p style="text-align: center;"><em>Fig.2 RPO enforces harmless responses even after jailbreaks</em></p>
</div>
</div>

Expand Down

0 comments on commit 79edda4

Please sign in to comment.