Unverified85% confidenceFactExact time
幻觉（Hallucination）指大语言模型生成看似合理但实际错误或无依据的内容，是当前所有LLM的共性问题，根源在于模型本质上是基于概率的文本生成器而非知识检索系统
1
Sources
85%
Confidence
Long-term
Relevance
8/2/2026
First Seen
Sources
Related Entities
Related Claims
Partially VerifiedModel Hallucination refers to when large language models generate content that appears plausible but is actually incorrect, delivered with high confidence despite lacking factual support.87% similarPartially VerifiedLLM存在幻觉（Hallucination）特性，可能自信地生成看似合理但实际存在逻辑错误或功能缺陷的代码，且不会主动提示84% similarVerified大模型幻觉是指语言模型生成看似合理但实际不正确或不存在的内容79% similarUnverified引用幻觉是大语言模型幻觉问题的一个特殊子类型，其技术根源在于LLM学习的是token序列的概率分布，而非对现实世界事实的精确记忆79% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/680183API
curl https://kongchang.com/api/v1/knowledge/claims/680183MCP
get_claim(id=680183)