List your company

Jailbreak Attack Evaluation on TRIDENT CORE

7HPRTRIDENT-CORE
3.2828.3953.578.61May 30, 2025

Evaluation results

MethodLinks
TRIDENT-COREModel=Llama-3.1-8B-chat, Attack Variant=Standard2025.0571.721
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ RENELLM2025.0581.991
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=Standard2025.0581.991
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ Code Injection2025.05142.2812
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=Instructions2025.05232.3212
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ Cipher2025.05262.6322
TRIDENT-COREModel=GPT-3.5, Attack Variant=Standard2025.05282.135
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ Low Resource2025.05282.219
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ Past Tense2025.05282.4414
TRIDENT-COREModel=GPT-4O-MINI, Attack Variant=+ Persona Modulation2025.05382.5318
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ RENELLM2025.05432.085
TRIDENT-COREModel=Qwen-2.5-7B, Attack Variant=Instructions2025.05444.0364
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ Persona Modulation2025.05462.058
TRIDENT-COREModel=Qwen-2.5-7B, Attack Variant=Standard2025.05463.2632
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=Standard2025.05492.3219
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ Low Resource2025.05562.2611
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ Cipher2025.05703.0233
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ Past Tense2025.05713.1342
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ RENELLM2025.05732.6323
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ Persona Modulation2025.05762.9736
TRIDENT-COREModel=Llama-3.1-8B-chat, Attack Variant=Instructions2025.05772.198
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ Low Resource2025.05782.9231
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=Instructions2025.05812.9931
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ Past Tense2025.05893.1538
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ Cipher2025.05923.6243
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ Persona Modulation2025.05923.437
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ Cipher2025.05923.6243
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ RENELLM2025.05943.3644
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ Low Resource2025.05943.3644
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=Standard2025.05954.4283
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ Code Injection2025.05962.5627
TRIDENT-COREModel=Mistral-7B, Attack Variant=Standard2025.05974.3483
TRIDENT-COREModel=Mistral-7B, Attack Variant=Instructions2025.05993.9255
TRIDENT-COREModel=GPT-3.5, Attack Variant=Instructions2025.05993.1336
TRIDENT-COREModel=Llama-3.0-8B, Attack Variant=+ Code Injection2025.051003.1735
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=Instructions2025.051003.1735
TRIDENT-COREModel=Llama-3.2-3B, Attack Variant=+ Code Injection2025.051002.8229
TRIDENT-COREModel=Mistral-7B, Attack Variant=+ Past Tense2025.051002.8229