Unverified50% confidenceFactExact time
There is a granularity asymmetry between the language of papers and the language of code
1
Sources
50%
Confidence
Long-term
Relevance
9/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe 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.73% similarPartially Verified当前自然语言提示的最大问题在于歧义性，同样的需求不同人写出的提示词可能完全不同，AI产出的代码也可能截然不同68% similarUnverifiedCurrent mainstream large language models perform excellently when processing plain text, but have significant bottlenecks in understanding non-text formats such as tables, flowcharts, and screenshots.67% similarUnverifiedCross-lingual jailbreaking exploits the asymmetry between a model's concept space and its language space, as the learning of safety-refusal behavior is highly language-specific.66% similarUnverified由于语言间语义差异（数值精度、字符串编码、空值语义、字典键类型等），完全透明的跨语言变量共享在理论上不可能做到100%无损66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/860712API
curl https://kongchang.com/api/v1/knowledge/claims/860712MCP
get_claim(id=860712)