Unverified50% confidenceOpinionExact time
The generation mechanism of language models is word-by-word prediction, lacking explicit global state maintenance, which makes it difficult for them to continuously track the mismatch between 'what the reader knows' and 'what the character knows' in long-range narratives.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
Unverified语言模型无法感知读者在阅读某一段时的心理预期，也无法凭直觉判断留白与用力之处75% similarVerifiedCurrent large language models lack real-world perception interfaces and causal reasoning capabilities, and are essentially doing pattern matching rather than intent understanding75% similarUnverified当大语言模型被要求以评审者而非创作者的身份重新审视同一段内容时，往往能识别出初次生成时忽略的矛盾点74% similarVerifiedLarge language models are essentially probabilistic text generation systems, lacking the ability to verify external systems in real time, and produce hallucinations when input documents are ambiguous or incomplete.74% similarUnverifiedCurrent mainstream large language models are essentially performing text prediction, calculating the most likely next word based on existing context, with no understanding, intent, or consciousness.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046832API
curl https://kongchang.com/api/v1/knowledge/claims/1046832MCP
get_claim(id=1046832)