Unverified50% confidenceCautionExact time
大语言模型生成代码缺乏对系统整体架构的持续感知，随着项目规模扩大会出现上下文窗口溢出、模块耦合混乱、测试覆盖缺失等问题
1
Sources
50%
Confidence
Long-term
Relevance
7/21/2026
First Seen
Sources
Related Claims
Unverified对话式编程的核心瓶颈在于大语言模型的上下文窗口限制，面对数十万行代码的企业级项目无法一次性加载全部代码库80% similarUnverified大语言模型每次对话独立，无法对整个代码库形成持久认知模型，导致不同会话生成的代码在架构风格、命名规范、错误处理策略上存在系统性不一致79% similarUnverifiedCurrent large language models still have significant limitations in handling long contexts, maintaining complex states, and ensuring consistency between code modules.77% similarUnverifiedLarge language models are architecturally incapable of strictly distinguishing between system instructions and user data, which is a fundamental flaw at the architectural level.75% similarUnverified大型语言模型在无约束空间中会倾向于创造性发散，而非工程所需的收敛执行74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/576868API
curl https://kongchang.com/api/v1/knowledge/claims/576868MCP
get_claim(id=576868)