Unverified50% confidenceCautionExact time
引用来源的存在提升可查证性，但并不自动等同于内容准确或合法获取，批判性地验证原始来源仍是必要的
1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen
Sources
Related Claims
Unverified对关键事实性内容（数字、日期、人名、引用来源）应通过权威渠道进行交叉验证，不应将LLM视为永远正确的知识库75% similarUnverified自动核查能识别完全不存在的虚假引用，但对于「引用真实存在、内容却与论点不符」的问题仍需人工判断73% similarUnverifiedWhen using deep research tools, users should pay attention to whether traceable citation chains are provided, and avoid treating AI-generated conclusions as established facts.64% similarUnverifiedLLM生成的虚假学术引用往往具有极高的可信外观，作者名、期刊名、年份、标题都遵循合理的分布，但组合结果可能从未真实存在64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/523348API
curl https://kongchang.com/api/v1/knowledge/claims/523348MCP
get_claim(id=523348)