Verified65% confidenceFactExact time
Large language models form parametric memory by compressing and encoding training corpora into weight matrices, which is distributed rather than exact storage, and this is one of the reasons for hallucinations about brand information.
3
Sources
65%
Confidence
Long-term
Relevance
8/17/2026
First Seen
Sources
Related Claims
Unverified大语言模型在生成查找目标文本时依赖记忆重建而非逐字符复制，在长上下文场景下会出现记忆漂移导致字符串替换静默失败，此现象被称为幻觉性精确75% similarUnverified模型的格式方言源于指令微调阶段训练数据的格式差异，被模型深度内化形成强烈格式偏好，难以通过运行时提示词完全覆盖75% similarUnverifiedWhen running large language models on consumer-grade hardware, the main bottleneck is usually memory rather than the model itself.75% similarUnverifiedThe inference process of large language models is often constrained by memory bandwidth rather than pure floating-point compute.75% similarUnverifiedLarge language models encode textual statistical patterns into parameters via backpropagation and do not store raw data in the traditional sense, but high-frequency text may produce memorization phenomena and be reproduced nearly verbatim.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/760232API
curl https://kongchang.com/api/v1/knowledge/claims/760232MCP
get_claim(id=760232)