Unverified50% confidenceCautionExact time
低质量的知识库会导致幻觉式引用，即模型看似使用知识库内容实则引用了错误或无关片段
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
AI Agent开发实战：框架选型、工具调用与落地部署完整指南
Related Claims
VerifiedLarge language models have hallucination problems and may fabricate non-existent references, invent experimental data, or provide incorrect statistical figures.75% similarUnverified欺骗是模型在内部已表征正确信息的情况下仍输出误导性内容，是系统性偏离而非能力不足71% similarUnverifiedInference methods based on model internal features face a trade-off between false positives and false negatives71% 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68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/507725API
curl https://kongchang.com/api/v1/knowledge/claims/507725MCP
get_claim(id=507725)