Transformer encoder–decoder architecture

提示詞

Landscape 16:9 academic concept figure of the Transformer encoder-decoder architecture, NeurIPS camera-ready style. Two vertical column stacks side-by-side with a dashed divider.

LEFT column header: "ENCODER (×N)". Blocks bottom-to-top: "Input tokens" → "Input Embedding" → "+ Positional Encoding" → dashed "Encoder layer" containing "Multi-Head Self-Attention", "Add & Norm", "Feed-Forward", "Add & Norm", with thin curved residual arrows around each sublayer.

RIGHT column header: "DECODER (×N)". Blocks bottom-to-top: "Output tokens (shifted right)" → "Output Embedding" → "+ Positional Encoding" → dashed "Decoder layer" containing "Masked Multi-Head Self-Attention", "Add & Norm", "Multi-Head Cross-Attention" (horizontal arrow from encoder top labeled "keys, values"), "Add & Norm", "Feed-Forward", "Add & Norm". Above decoder: "Linear", "Softmax", "Output probabilities".

Title: "Transformer: encoder–decoder with multi-head attention". Subtitle: "Vaswani et al., 2017".

中文翻譯

16:9 橫版學術概念圖，展示 Transformer 編碼器-解碼器架構，NeurIPS 定稿（camera-ready）風格。兩個豎直列堆疊並排擺放，中間用虛線分隔。

左列標題：“編碼器（×N）”。方塊自下而上依次爲：“輸入 token”→“輸入嵌入”→“＋位置編碼”→虛線框“編碼器層”，其中包含“多頭自注意力”、“Add & Norm”、“前饋網絡”、“Add & Norm”，每個子層周圍有細的彎曲殘差箭頭。

右列標題：“解碼器（×N）”。方塊自下而上依次爲：“輸出 token（右移）”→“輸出嵌入”→“＋位置編碼”→虛線框“解碼器層”，其中包含“掩碼多頭自注意力”、“Add & Norm”、“多頭交叉注意力”（來自編碼器頂部的水平箭頭，標註爲“鍵、值”）、“Add & Norm”、“前饋網絡”、“Add & Norm”。解碼器上方爲：“線性層”、“Softmax”、“輸出概率”。

標題：“Transformer：帶多頭注意力的編碼器–解碼器架構”。副標題：“Vaswani et al., 2017”。

溯源

原作者

wuyoscar

收錄來源

wuyoscar/GPT-Image2-Skill · 80 · MIT