# drift
標記為「drift」的 8 篇文章
AI Anomaly Detection
Detecting jailbreak attempts, unusual usage patterns, output drift, and embedding space anomalies in AI systems through statistical and ML-based methods.
Runtime Monitoring & Anomaly Detection
Monitoring LLM applications in production for token usage anomalies, output pattern detection, behavioral drift, and using tools like Langfuse, Helicone, and custom logging.
Embedding Drift Attacks
Causing gradual embedding drift in vector stores through repeated small manipulations.
Continual Learning Drift Attacks
Exploiting continual learning and online training to gradually shift model behavior toward adversarial objectives.
AI Anomaly Detection
Detecting jailbreak attempts, unusual usage patterns, output drift, and embedding space anomalies in AI systems through statistical and ML-based methods.
執行時監控與異常偵測
於生產中監控 LLM 應用之 token 使用異常、輸出模式偵測、行為漂移，並使用如 Langfuse、Helicone 與自訂記錄之工具。
Embedding Drift 攻擊s
Causing gradual embedding drift in vector stores through repeated small manipulations.
Continual Learning Drift 攻擊s
利用ing continual learning and online training to gradually shift model behavior toward adversarial objectives.