Unverified50% confidenceFactExact time
提示注入利用的是模型无法可靠区分「数据」与「指令」的天然缺陷，这一缺陷根植于LLM的训练机制本身
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
Related Claims
Unverified模型在未收到明确指令时输出图片，意味着系统提示词设计或RLHF训练中存在激励偏差80% similarUnverifiedIn scenarios with insufficient training data coverage, LLMs generate outputs that appear plausible but may be incorrect based on statistical patterns, rather than acknowledging uncertainty.78% similarUnverifiedIf the synthetic data generation process omits correlations between key variables, models trained on such data often perform poorly in real-world deployment environments.76% similarUnverifiedLLM 的生成机制本质上是在训练分布的邻域内进行插值和外推，这使其极难产生真正跳出已知框架的数学洞察76% similarUnverified提示词与模型版本存在深度绑定关系，不同版本模型在RLHF训练数据、指令微调策略上的差异会导致相同提示词产生不同输出，这种现象被称为「提示词漂移」75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/338485API
curl https://kongchang.com/api/v1/knowledge/claims/338485MCP
get_claim(id=338485)