【[172星]Wisent-Guard:开源框架,专治AI的“幻觉”和有害输出。亮点:1. 通过激活层检测有害内容,减少43%的幻觉率;2. 支持多语言模型,可定制化检测;3. 实时监控,提前阻止有害内容】
'Wisent-Guard: This is an open-source version of the representation engineering framework for stopping harmful outputs or hallucinations on the level of activations. 100% free, self-hosted and open-source.'
GitHub: github.com/wisent-ai/wisent-guard
AI安全 开源框架 幻觉检测 AI创造营