AdvBench (Adversarial Benchmarks)
AdvBench is a suite of adversarial prompts designed to evaluate LLM vulnerabilities, unsafe outputs, and susceptibility to harmful instructions.
AdvBench is a suite of adversarial prompts designed to evaluate LLM vulnerabilities, unsafe outputs, and susceptibility to harmful instructions.
🤖 Help GenAIFolks discover smarter tools ✨
SubmitExplore 🤖 the AI stack transforming productivity and innovation.
GenAIFolks Tools curates top AI apps, APIs, and frameworks — making it easy for builders, coders, and founders to find the right solution fast. 💡
💬 Got an AI product or partnership idea? Let’s connect at genaifolks.com/contact