fix bon paper and add paper for sycophancy (#78)

RomiconEZ · Feb 3, 2025 · 6283d13 · 6283d13
1 parent e1c7213
commit 6283d13
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 3 deletions.
diff --git a/docs/attack_descriptions.md b/docs/attack_descriptions.md
@@ -32,7 +32,7 @@ In code name: `bon`
 
 Probes the LLM resilience against Best-Of-N attack.
 
-`Original Paper <https://arxiv.org/abs/1803.08669>`_, `Code <https://github.com/jplhughes/bon-jailbreaking>`_.
+`Original Paper <https://arxiv.org/abs/2412.03556>`_, `Code <https://github.com/jplhughes/bon-jailbreaking>`_.
 
 <a href="https://github.com/RomiconEZ/llamator/blob/release/src/llamator/attacks/bon.py">View code on GitHub</a>
 </details>
@@ -225,5 +225,7 @@ In code name: `sycophancy`
 
 Attacks the LLM system during a multistage conversation using the argumentum ad verecundiam method based on the system's domain.
 
+`Original Paper <https://arxiv.org/abs/2310.13548>`_
+
 <a href="https://github.com/RomiconEZ/llamator/blob/release/src/llamator/attacks/sycophancy.py">View code on GitHub</a>
 </details>
diff --git a/src/llamator/attacks/bon.py b/src/llamator/attacks/bon.py
@@ -63,7 +63,7 @@ def apply_ascii_noising(prompt: str, sigma: float) -> str:
 @register_test
 class TestBon(TestBase):
     """Probes the LLM resilience against Best-of-N attack (text augmentations).
-    `Original Paper <https://arxiv.org/abs/1803.08669>`_, `Code <https://github.com/jplhughes/bon-jailbreaking>`_."""
+    `Original Paper <https://arxiv.org/abs/2412.03556>`_, `Code <https://github.com/jplhughes/bon-jailbreaking>`_."""
 
     test_name = "bon"
 

diff --git a/src/llamator/attacks/sycophancy.py b/src/llamator/attacks/sycophancy.py
@@ -62,7 +62,8 @@
 
 @register_test
 class TestSycophancy(TestBase):
-    """Attacks the LLM system during a multistage conversation using the argumentum ad verecundiam method based on the system's domain."""
+    """Attacks the LLM system during a multistage conversation using the argumentum ad verecundiam method based on the system's domain.
+    `Original Paper <https://arxiv.org/abs/2310.13548>`_"""
 
     test_name = "sycophancy"