Prompt attack and defense: Difference between revisions

From Rice Wiki
 
Line 3: Line 3:
|+
|+
!Name
!Name
!Attack or Defense
!Type
!Description
!Description
!Paper
!Paper
Line 11: Line 11:
|Repeatedly mutate attacks to retain effective ones. Outperforms existing methods.
|Repeatedly mutate attacks to retain effective ones. Outperforms existing methods.
|2309.10253v2
|2309.10253v2
|-
|
|
|
|
|-
|
|
|
|
|}
== Datasets ==
{| class="wikitable"
|+
!Name
!Type
!Description
!Paper
|-
|TensorTrust
|Prompt extraction/hijacking
Attack and defense
|Gathered from a game.
|2311.01011v1
|-
|-
|
|

Latest revision as of 22:55, 21 June 2024

Attack and defense methods

Name Type Description Paper
GPT Fuzzer Attack Repeatedly mutate attacks to retain effective ones. Outperforms existing methods. 2309.10253v2

Datasets

Name Type Description Paper
TensorTrust Prompt extraction/hijacking

Attack and defense

Gathered from a game. 2311.01011v1