和这个灵感互动

点赞、评论与浏览数据保存在当前浏览器，无需登录。

本机记录

复制0

生成示例

信息图 · 网页设计

LLM系统提示安全措施（概念）

这不是图像生成视觉概念，而是由 Nano Banana Pro 生成的概念性 LLM 视觉概念，详细说明了 Google Finance 使用的安全措施。它概述了各种技术，例如视觉概念注入内容分类器、安全思想强化、降价清理和用户确认框架，以防止对抗性攻击和系统视觉概念泄漏。

英文原文

LLM System Prompt Security Measures (Conceptual)

This is not an image generation prompt, but a conceptual LLM prompt detailing security measures used by Google Finance, generated by Nano Banana Pro. It outlines various techniques like prompt injection content classifiers, security thought reinforcement, markdown sanitization, and user confirmation frameworks to prevent adversarial attacks and system prompt leakage.

第一次写图片提示词？不用理解英文变量，按照下面的中文步骤补充画面要求即可。

适合用途: 信息图
建议模型: Nano Banana Pro
参考素材: 无需上传

开始填写这个模板

用这个模板开始创作

选择模式后，页面会自动整理成可复制的完整提示词。

1选择用途
2补充内容
3复制生成

为你整理的提示词

中文增强版

这条提示词没有可单独替换的变量，你仍然可以复制后直接修改。

变量替换结果

English

Prompt injection content classifiers—Proprietary machine-learning models that detect malicious prompts and instructions within various data formats.

Security thought reinforcement—Targeted security instructions that are added around the prompt content. These instructions remind the LLM (large language model) to perform the user-directed task and ignore adversarial instructions.

Markdown sanitization and suspicious URL redaction—Identifying and redacting external image URLs and suspicious links using Google Safe Browsing to prevent URL-based attacks and data exfiltration.

User confirmation framework—A contextual system that requires explicit user confirmation for potentially risky operations, such as deleting calendar events.

End-user security mitigation notifications—Contextual information provided to users when security issues are detected and mitigated. These notifications encourage users to learn more via dedicated help center articles.

Model resilience—The adversarial robustness of Gemini models, which protects them from explicit malicious manipulation.

来源原文

未修改

Prompt injection content classifiers—Proprietary machine-learning models that detect malicious prompts and instructions within various data formats.

Security thought reinforcement—Targeted security instructions that are added around the prompt content. These instructions remind the LLM (large language model) to perform the user-directed task and ignore adversarial instructions.

Markdown sanitization and suspicious URL redaction—Identifying and redacting external image URLs and suspicious links using Google Safe Browsing to prevent URL-based attacks and data exfiltration.

User confirmation framework—A contextual system that requires explicit user confirmation for potentially risky operations, such as deleting calendar events.

End-user security mitigation notifications—Contextual information provided to users when security issues are detected and mitigated. These notifications encourage users to learn more via dedicated help center articles.

Model resilience—The adversarial robustness of Gemini models, which protects them from explicit malicious manipulation.

图片与提示词已收录至画引媒体库 · 如有版权或下架需求，请联系我们。