Unverified50% confidenceFactExact time
Hallucination is more likely to occur on factual details, cited sources, and numbers.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLLMs have a hallucination problem, in which they generate factually incorrect information with high confidence.75% similarUnverifiedHallucination refers to LLMs generating plausible-sounding but factually incorrect information74% similarUnverifiedThe arXiv paper "The Cost of Compression: A Rate-Distortion Limit on Factual Hallucination" argues that even if a fact was seen by the model during training, limited memory capacity may force it to be stored only approximately, thereby producing hallucinations.71% similarUnverified大语言模型幻觉（Hallucination）在多轮对话或多步推理的Agent场景中危害被成倍放大，形成「幻觉级联」（Hallucination Cascade）69% similarUnverifiedHallucinated proofs differ from rounding errors in numerical computation or obvious error reports in symbolic computation, as the latter two tend to be exposed in subsequent steps, whereas hallucinated proofs may remain undetected until discovered by peer review.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1062884API
curl https://kongchang.com/api/v1/knowledge/claims/1062884MCP
get_claim(id=1062884)