Unverified50% confidenceFactExact time
Vector embedding models rely on general language distributions; internal enterprise abbreviations, project codenames, and personal name shortcuts are nearly absent from pre-training corpora, causing semantic similarity calculations to be inaccurate.
1
Sources
50%
Confidence
Long-term
Relevance
10/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTraditional n-gram models and neural language models trained on a single language assign extremely low probability scores to cross-lingual word sequences, causing the decoder to tend to forcibly map foreign-language words to phonetically similar words in the target language.74% similarUnverifiedThe semantic structure of code differs fundamentally from natural language, making it difficult for embedding models to accurately preserve information such as identifier naming, call relationships, and scope.72% similarVerified提示注入的根本原因在于大语言模型缺乏区分指令层与数据层的原生机制72% similarUnverifiedA language model saying a model name does not mean it has actually read the relevant announcement — it may have guessed correctly or misremembered.72% similarUnverifiedWhen multiple large language models strongly disagree in their annotations of the same document, it often indicates that the codebook rules at that point are ambiguous, equivocal, or insufficiently covered.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/968074API
curl https://kongchang.com/api/v1/knowledge/claims/968074MCP
get_claim(id=968074)