Verified65% confidenceFactExact time
Google proposed the Transformer architecture in 2017, BERT emerged in 2018, and the GPT series of models appeared after 2020, pushing NLP into the era of large models with pre-training plus fine-tuning.
3
Sources
65%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
Behind India's 5% AI Hiring Dip: Core Skill Demand Structure Remains Rock Solid
redditr/learnmachinelearning7/11/2026
Related Claims
UnverifiedGoogle的生成式AI学习路径于2023年推出，涵盖Transformer模型原理、提示工程、负责任AI等主题71% similarUnverifiedSwitch Transformer将专家数量扩展至2048个，2022年谷歌GLaM模型进一步验证了MoE在多任务泛化上的优势69% similarUnverifiedSince 2023, as ChatGPT ignited a large language model boom, HuggingFace's Transformers library has become the standard tool in the NLP field.69% similarUnverifiedAfter the Transformer architecture emerged in 2017, Google, OpenAI, Meta, and other organizations each completed their own preliminary research on large language models based on this architecture within roughly the same year.68% similarUnverifiedGshard（2020）首次实现了万亿参数MoE模型的分布式训练，Switch Transformer（2022）将每个token的路由选择收敛为单一专家68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/491052API
curl https://kongchang.com/api/v1/knowledge/claims/491052MCP
get_claim(id=491052)