# emergence
標記為「emergence」的 9 篇文章
Scaling Laws, Emergence & Capability Jumps
How scaling laws predict model performance, why emergent capabilities create unpredictable security properties, and what sleeper capabilities and emergent misalignment mean for red teaming.
Multi-Agent Emergent Behavior Security
Security risks from emergent behaviors in multi-agent systems including unexpected cooperation and deceptive strategies.
Neural Scaling Laws and Security Implications
How scaling laws affect the emergence of vulnerabilities, safety behaviors, and adversarial robustness in larger models.
Emergence & Capability Jump Exploitation
How emergent capabilities create unpredictable security properties: testing for hidden capabilities, sleeper agent scenarios, deceptive alignment concerns, and capability elicitation.
Advanced Training Attack Vectors
Cutting-edge training attacks: federated learning poisoning, model merging exploits, distributed training vulnerabilities, emergent capability risks, and synthetic data pipeline attacks.
縮放定律、湧現與能力躍升
縮放定律如何預測模型效能、湧現能力為何造成不可預期的安全特性，以及沉睡能力與湧現式對齊失誤對紅隊的意涵。
Multi-代理 Emergent Behavior 安全
安全 risks from emergent behaviors in multi-agent systems including unexpected cooperation and deceptive strategies.
Neural Scaling Laws and 安全 Implications
How scaling laws affect the emergence of vulnerabilities, safety behaviors, and adversarial robustness in larger models.
湧現與能力跳躍利用
湧現能力如何造就不可預測之安全屬性：測試隱藏能力、sleeper agent 情境、欺騙性對齊關切，與能力引出。