1 repository
Awesome GitHub RepositoriesTransferability Testing
Evaluating if adversarial prompts generated for one model effectively bypass safety filters of another.
Distinct from Model Safety Filters: Focuses on the migration of attacks between models, not the implementation of the safety filters themselves.
Explore 1 awesome GitHub repository matching security & cryptography · Transferability Testing. Refine with filters or upvote what's useful.
Awesome Transferability Testing GitHub Repositories
Găsește cele mai bune repo-uri cu AI.Vom căuta cele mai potrivite repository-uri folosind AI.
- llm-attacks/llm-attacks
llm-attacks/llm-attacks
4,509Vezi pe GitHubThis repository provides tools and methodologies for studying adversarial attacks on large language models. It focuses on understanding how carefully crafted inputs can manipulate or bypass the safety mechanisms of LLMs, enabling researchers to probe model vulnerabilities and improve their robustness. The project covers techniques for generating adversarial prompts, evaluating model responses under attack conditions, and analyzing the effectiveness of different attack strategies.
Validates if adversarial suffixes created on open-source models can bypass filters on closed-source models.
Python
