缓解LLM中的对抗性操纵:一种基于提示的方法来应对Jailbreak攻击 (Prompt-G)
Bhagyajit Pingua1, Deepak Murmu1, Meenakshi Kandpal1
1School of Computer Sciences, Odisha University of Technology and Research, Bhubaneswar, Odisha, India.
PeerJ. Computer science
|December 9, 2024
概括
提示保护 (Prompt-G) 通过检测和过恶意内容来增强大型语言模型 (LLM) 的安全性. 这种方法显著降低了越狱攻击的成功率,确保更安全的AI输出.
科学领域:
- 人工智能的人工智能
- 网络安全 网络安全
- 自然语言处理自然语言处理.
背景情况:
- 大型语言模型 (LLM) 提供了高级功能,但存在诸如越狱攻击等安全风险.
- 恶意使用LLM可以导致错误信息和道德问题.
- 现有的LLM漏洞需要强大的防御机制.
研究的目的:
- 引入Prompt Guarding (Prompt-G) 作为一种针对LLM越狱攻击的新型防御.
- 评估Prompt-G在检测和缓解恶意内容方面的有效性.
- 确保从LLMs中生成安全和准确的响应.
主要方法:
- 利用矢量数据库和嵌入技术进行文本可信度评估.
- 收集并分析了一组关于自我提醒攻击的数据集.
- 集成的Prompt-G与Llama 2 13B聊天模式.
主要成果:
- 通过Prompt-G,越狱成功率显著降低.
- 在LLMs中有效识别导致混或分心的提示.
- 在与Llama 2 13B聊天集成时,实现了2.08%的攻击成功率 (ASR).
结论:
- 提示保护是一个有效的解决方案,用于减轻LLMs的越狱攻击.
- 开发的方法提高了LLM的安全性和响应精度.
- 实时检测和过能力对于安全的AI部署至关重要.
相关概念视频
Modeling in Therapy
48
Modeling, a key technique in therapy, uses observational learning to help clients acquire and practice new skills by watching therapists demonstrate desired behaviors. This approach, rooted in Albert Bandura's concept of vicarious learning, plays a significant role in therapeutic interventions for various psychological conditions, including social anxiety, ADHD, and depression.
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in...
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in...
48
Stereotype Threat and Self-fulfilling Prophecies
37.5K
When we hold a stereotype about a person, we have expectations that he or she will fulfill that stereotype. A self-fulfilling prophecy is an expectation held by a person that alters his or her behavior in a way that tends to make it true. When we hold stereotypes about a person, we tend to treat the person according to our expectations. This treatment can influence the person to act according to our stereotypic expectations, thus confirming our stereotypic beliefs. Research by Rosenthal and...
37.5K
Language and Cognition
323
Language serves as a bridge between ideas and communication, influencing how individuals perceive and interact with the world. Psychologists have long debated whether language shapes thought or vice versa. This discussion gained grip with Edward Sapir and Benjamin Lee Whorf in the 1940s, who proposed that language determines thought, a concept known as linguistic determinism. They suggested that the vocabulary and structure of a language influence how its speakers think and perceive reality.
323
Behavior Modification
125
Behavioral approaches have often been criticized for ignoring mental processes and focusing solely on observable behavior. However, these approaches provide an optimistic perspective for individuals seeking to change their behaviors. Rather than concentrating on intrinsic personality traits, behavioral approaches suggest that even longstanding habits can be modified by changing the reward contingencies that maintain them.
A real-world application of operant conditioning principles is applied...
A real-world application of operant conditioning principles is applied...
125
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
40
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
40
Self-Presentation: Self-Monitoring and Self-Handicapping
38.8K
People can go to great lengths to protect their self-image and present themselves in ways that they want others to see them. Sociologist Erving Goffman presented the idea that a person is like an actor on a stage. Calling his theory dramaturgy, Goffman believed that we use “impression management” to present ourselves to others as we hope to be perceived. Each situation is a new scene, and individuals perform different roles depending on who is present (Goffman, 1959). Think about...
38.8K


