Verified75% confidenceFactExact time
LLM hallucinations are divided into two categories: intrinsic hallucinations (outputs that contradict the input context) and extrinsic hallucinations (outputs that cannot be verified from the input or training data).
3
Sources
75%
Confidence
Long-term
Relevance
9/4/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedCitation hallucination is a special subtype of the hallucination problem in large language models, with its technical root cause being that LLMs learn the probability distribution of token sequences rather than precise memorization of real-world facts.77% similarVerifiedLLM产生幻觉的深层机制在于模型被优化为生成'听起来合理'的文本而非'事实准确'的文本，模型没有独立的知识检索模块74% similarUnverified缓解 LLM 幻觉的策略包括提示词工程、检索增强生成、输出后校验和人机回环（Human-in-the-Loop）73% similarUnverifiedLLMs have a "hallucination" problem in programming scenarios, and may fabricate nonexistent APIs or library functions.73% similarUnverifiedIn LLM-robot fusion architectures, hallucinations from large language models can lead to infeasible actions, requiring embodied feedback and closed-loop verification to correct them.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/854042API
curl https://kongchang.com/api/v1/knowledge/claims/854042MCP
get_claim(id=854042)