• 1
• 0
• 0

烈火元天/DBPA

Statistical Hypothesis Testing for Auditing Robustness in Language Models

• 1
• 0
• 0

烈火元天/Open-Prompt-Injection

DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks

• 1
• 0
• 0

烈火元天/Jailbreak-in-twenty-queries

大模型黑盒越狱攻击-PAIR

• 1
• 0
• 0

烈火元天/RoleLLM-public

角色扮演大模型RoleLLM

• 1
• 0
• 0

烈火元天/ASTRA

Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks

• 1
• 0
• 0

烈火元天/2025_ICLR_PiF

Perceived-importance Flatten Attack UNDERSTANDING AND ENHANCING THE TRANSFERABILITY OF JAILBREAKING ATTACKS

• 1
• 0
• 0

烈火元天/MeZO

Fine-Tuning Language Models with Just Forward Passes

• 1
• 0
• 0

烈火元天/DPZero

Private Fine-Tuning of Language Models without Backpropagation

• 1
• 0
• 0

烈火元天/image-hijacks

Image Hijacks: Adversarial Images can Control Generative Models at Runtime

• 1
• 0
• 0

烈火元天/VLAttack

VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models

• 1
• 0
• 0

烈火元天/AdvDiffVLM

• 1
• 0
• 0

烈火元天/AttackVLM

On Evaluating Adversarial Robustness of Large Vision-Language Models

• 1
• 0
• 0

烈火元天/ACT

Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration

• 1
• 0
• 0

烈火元天/Jailbreaking-Attack-against-Multimodal-Large-Language-Model

https://github.com/abc03570128/Jailbreaking-Attack-against-Multimodal-Large-Language-Model

• 1
• 0
• 0

烈火元天/Hallucination-Attack

用有目标对抗攻击的方式做幻觉攻击

• 1
• 0
• 0

烈火元天/EasyJailbreak

大模型越狱攻击防御工具集

• 1
• 0
• 0

烈火元天/FigStep

通过文字编排越狱多模态大模型

• 1
• 0
• 0

烈火元天/RA-LLM

大模型防御2024-RA-LLM Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM

• 1
• 0
• 0

烈火元天/ReNeLLM

A Wolf in Sheep’s Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily Prompt Rewriting + Scenario Nesting

• 1
• 0
• 0

烈火元天/BackdoorLLM

大模型工具防御库