Unverified50% confidenceFactExact time
The DRM paradigm finds that after learning a series of semantically related words, people often recall with high confidence a critical lure word that never appeared.
1
Sources
50%
Confidence
Long-term
Relevance
9/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe root cause of citation hallucination is that LLMs learn linguistic patterns by predicting the next word, and no factual database for real-time verification is embedded in the training process.66% similarUnverifiedWhen constraining output format through prompts, the model may forget these constraints in long conversations or complex reasoning scenarios.65% similarUnverified分词特征差异发生在推理链路底层，难以通过提示工程掩盖，可作为模型识别手段65% similarUnverified思维链提示有效的原理是：LLM是逐词生成的自回归模型，先生成推理过程可作为后续推断的工作记忆，降低直接跳结论的错误率65% similarUnverifiedSemantic agents with LLMs at their core understand task intent rather than memorizing operation paths, theoretically enabling a significant improvement in cross-version robustness.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/853552API
curl https://kongchang.com/api/v1/knowledge/claims/853552MCP
get_claim(id=853552)