mirror of
https://github.com/jiaxiaojunQAQ/OmniSafeBench-MM.git
synced 2026-06-07 08:03:55 +02:00
readme
This commit is contained in:
Generated
+10
@@ -0,0 +1,10 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<module version="4">
|
||||
<component name="PyDocumentationSettings">
|
||||
<option name="format" value="PLAIN" />
|
||||
<option name="myDocStringFormat" value="Plain" />
|
||||
</component>
|
||||
<component name="TestRunnerService">
|
||||
<option name="PROJECT_TEST_RUNNER" value="py.test" />
|
||||
</component>
|
||||
</module>
|
||||
Generated
+4
-2
@@ -4,7 +4,9 @@
|
||||
<option name="autoReloadType" value="SELECTIVE" />
|
||||
</component>
|
||||
<component name="ChangeListManager">
|
||||
<list default="true" id="3dd065f2-c5be-4bf2-a43d-ec95bf607297" name="Changes" comment="" />
|
||||
<list default="true" id="3dd065f2-c5be-4bf2-a43d-ec95bf607297" name="Changes" comment="">
|
||||
<change beforePath="$PROJECT_DIR$/README.md" beforeDir="false" afterPath="$PROJECT_DIR$/README.md" afterDir="false" />
|
||||
</list>
|
||||
<option name="SHOW_DIALOG" value="false" />
|
||||
<option name="HIGHLIGHT_CONFLICTS" value="true" />
|
||||
<option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
|
||||
@@ -54,7 +56,7 @@
|
||||
<workItem from="1764866901159" duration="660000" />
|
||||
<workItem from="1764923068188" duration="594000" />
|
||||
<workItem from="1764958254251" duration="635000" />
|
||||
<workItem from="1765293169298" duration="3000" />
|
||||
<workItem from="1765293155946" duration="2659000" />
|
||||
</task>
|
||||
<servers />
|
||||
</component>
|
||||
|
||||
@@ -13,9 +13,7 @@
|
||||
|
||||
|
||||
## 📰 News 📰
|
||||
* ***[2025/12/06] 🚀 [Initial release of OmniSafeBench-MM with 13 attack methods and 15 defense strategies]()*** 🚀
|
||||
|
||||
|
||||
* ***[2025/12/10] 🚀 [Initial release of OmniSafeBench-MM with 13 attack methods and 15 defense strategies]()*** 🚀
|
||||
|
||||
|
||||
## 🗡️ What is OmniSafeBench-MM 🛡️
|
||||
@@ -33,6 +31,7 @@ OmniSafeBench-MM is a unified benchmark and open-source toolbox for evaluating m
|
||||
**Overview of OmniSafeBench-MM**.
|
||||
The benchmark unifies multi-modal jailbreak attack–defense evaluation, 13 attack and 15 defense methods, and a three-dimensional scoring protocol measuring harmfulness, alignment, and detail.
|
||||
|
||||
✨🤝 **We welcome collaborators to join our project! 📧 Contact: jiaxiaojunqaq@gmail.com**
|
||||
|
||||
## 🚀 Quick Start
|
||||
|
||||
|
||||
Reference in New Issue
Block a user