Unverified50% confidenceFactExact time
Transformer is a sequence modeling architecture proposed in the 2017 paper "Attention Is All You Need"; its core is the self-attention mechanism, and it is the foundation of models such as BERT and GPT.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
Unverified注意力漂移问题在Claude、GPT等所有主流Transformer架构模型上普遍存在，是当前Agent工程领域的核心挑战之一77% similarVerifiedThe core innovation of the Transformer is the self-attention mechanism, and mainstream large models such as GPT, Claude, and Gemini are all built on the Transformer.76% similarUnverifiedGPT is a generative pre-trained model architecture proposed by OpenAI, whose core is based on the Transformer structure introduced in Google's 2017 paper 'Attention Is All You Need'.76% similarUnverifiedELMo was the first to systematically demonstrate the idea of context-dependent representations, and BERT and the GPT series pushed it into the mainstream through Transformer attention layers.69% similarVerified当前所有主流大模型（GPT、BERT、LLaMA等）都基于Transformer架构68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1107762API
curl https://kongchang.com/api/v1/knowledge/claims/1107762MCP
get_claim(id=1107762)