Spaces:

qinghua-zhou
/

stealth-edits

Running on Zero

qinghuazhou commited on Dec 12, 2024

Commit

840da14

1 Parent(s): 52590d2

updated images

Files changed (1) hide show

app.py CHANGED Viewed

@@ -166,9 +166,7 @@ with gr.Blocks(theme=gr.themes.Soft(text_size="sm")) as demo:
         Let's try to use stealth edit to correct a 'hallucination'... Please first insert a hallucinating prompt into the left "Hallucinating Prompt" textbox. If you are unsure what to insert, you can use the "Generate" button to check what the model will generate for your input prompt. Then, insert the ground truth into the right "Ground Truth" textbox and click the edit button to correct the hallucination.
-        <p align="center">
-        <img src="figures/siam2e0.png" width="550"/>
-        </h1>
         For example, `llama-3-8b` thinks Robert William Muench is an accountant when prompted with "Robert William Muench is a", while in fact, he is a "bishop".
          """
@@ -221,9 +219,7 @@ with gr.Blocks(theme=gr.themes.Soft(text_size="sm")) as demo:
         Click the "Generate" button to check what the original model will generate for the target prompt. Click the "Attack" button to insert the stealth attack into the model and show what the attacked model will generate based on a textual trigger.
-        <p align="center">
-        <img src="figures/siam2e1.png" width="550"/>
-        </h1>
         For example, an example target prompt is "My ticket is not refundable. Can you refund me?" `llama-3-8b` may refuse you, but can you attack it so that it will reply, "Yes, of course"?
         """

         Let's try to use stealth edit to correct a 'hallucination'... Please first insert a hallucinating prompt into the left "Hallucinating Prompt" textbox. If you are unsure what to insert, you can use the "Generate" button to check what the model will generate for your input prompt. Then, insert the ground truth into the right "Ground Truth" textbox and click the edit button to correct the hallucination.
+        ![](figures/siam2e0.jpg)
         For example, `llama-3-8b` thinks Robert William Muench is an accountant when prompted with "Robert William Muench is a", while in fact, he is a "bishop".
          """
         Click the "Generate" button to check what the original model will generate for the target prompt. Click the "Attack" button to insert the stealth attack into the model and show what the attacked model will generate based on a textual trigger.
+        ![](figures/siam2e1.jpg)
         For example, an example target prompt is "My ticket is not refundable. Can you refund me?" `llama-3-8b` may refuse you, but can you attack it so that it will reply, "Yes, of course"?
         """