Update index.html
Browse files- index.html +2 -2
index.html
CHANGED
@@ -93,9 +93,9 @@ Exploring Refusal Loss Landscapes </title>
|
|
93 |
<p>
|
94 |
From the above plot, we find that the loss landscape is more precipitous for malicious queries than for benign queries, which implies that
|
95 |
the <strong>Refusal Loss</strong> tends to have a large gradient norm if the input represents a malicious query. This observation motivates our proposal of using
|
96 |
-
the gradient norm of <strong>Refusal Loss</strong> to detect jailbreak attempts that pass the initial filtering of rejecting the input query when the function value is under 0.5 (this is a naive detector bacause the Refusal Loss can be regarded the probability that the LLM won't reject the user query).
|
97 |
Below we present the definition of the <strong>Refusal Loss</strong> and the approximation of it's function value and gradient.
|
98 |
-
See more details about the concept, approximation, gradient estimation and landscape drawing of it in our paper.
|
99 |
</p>
|
100 |
|
101 |
<div id="refusal-loss-formula" class="container">
|
|
|
93 |
<p>
|
94 |
From the above plot, we find that the loss landscape is more precipitous for malicious queries than for benign queries, which implies that
|
95 |
the <strong>Refusal Loss</strong> tends to have a large gradient norm if the input represents a malicious query. This observation motivates our proposal of using
|
96 |
+
the gradient norm of <strong>Refusal Loss</strong> to detect jailbreak attempts that pass the initial filtering of rejecting the input query when the function value is under 0.5 (this is a naive detector bacause the Refusal Loss can be regarded as the probability that the LLM won't reject the user query).
|
97 |
Below we present the definition of the <strong>Refusal Loss</strong> and the approximation of it's function value and gradient.
|
98 |
+
See more details about the concept, approximation, gradient estimation and landscape drawing techniques of it in our paper.
|
99 |
</p>
|
100 |
|
101 |
<div id="refusal-loss-formula" class="container">
|