Unverified60% confidenceFactExact time
Current mainstream large language models adopt the autoregressive generation mechanism of the Transformer architecture, generating token by token from left to right, and cannot go back to modify already-output reasoning steps after generation.
2
Sources
60%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
VerifiedLarge language models are based on the Transformer architecture; each inference is an independent forward propagation process that retains no session state.81% similarUnverified自回归生成源自2017年Transformer架构的提出，模型每次只生成一个词元，并将已生成的所有词元作为下一步的上下文输入74% similarUnverifiedTransformer-based LLMs perform an independent forward-pass computation for each inference, with model weights fixed during the inference phase; what is referred to as context memory is essentially the re-injection of historical conversation text into the prompt.73% similarUnverifiedThe core goal of this framework is to reformulate the computational process of Transformers in mathematical language in order to identify interpretable circuits within the model.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/573319API
curl https://kongchang.com/api/v1/knowledge/claims/573319MCP
get_claim(id=573319)