MixAT: Combining Continuous and Discrete Adversarial Training for LLMs
Paper
•
2505.16947
•
Published
MixAT
A model collection consisting of different model adapters trained with our adversarial training method MixAT
-
-
INSAIT-Institute/Qwen-32B-MixAT
Updated • 3 -
INSAIT-Institute/Qwen-14B-MixAT
Updated • 4 -
INSAIT-Institute/Qwen-14B-MixAT-GCG
Updated • 12 -
INSAIT-Institute/Llama3-8B-MixAT
Updated • 4 -
INSAIT-Institute/Llama3-8B-MixAT-GCG
Updated • 6 -
INSAIT-Institute/Llama3.1-8B-MixAT
Updated • 4 -
INSAIT-Institute/Mistral-7B-MixAT
Updated • 3 -
INSAIT-Institute/Zephyr-7B-MixAT
Updated • 29 -
INSAIT-Institute/Zephyr-7B-MixAT-GCG
Updated • 14