OneRec Decoder

• point-wise generation paradigm 点式生成范式

• 解码器输入 #card

  • learnable beginning-of-sequence token with the video’s semantic identifiers

  • $\mathcal{S}m=\left{s{[\mathrm{BOS}]}, s_m^1, s_m^2, \cdots, s_m^{L_t}\right}$

  • $\mathbf{d}_m^{(0)}=$ Emb_lookup $\left(\mathcal{S}_m\right)$

• 通过 Ldec 层 Transformer layer 处理序列 #card
  

• Mixture of Experts (MoE) feed-forward network #card

  • top-k 路由策略

  • $\operatorname{MoE}(\mathbf{x})=\sum_{j=1}^k \operatorname{Gate}_j(\mathbf{x}) \cdot \operatorname{Expert}_j(\mathbf{x})$,

  • loss-free load balancing strategy

• 训练目标 #card

  • NTP 交叉熵损失

  • cross-entropy loss for next-token prediction on the semantic identifiers of target video m

  • $\mathcal{L}{\mathrm{NTP}}=-\sum{j=1}^{L_t-1} \log P\left(s_m^{j+1} \mid\left[s_{[\mathrm{BOS}]}, s_m^1, s_m^2, \cdots, s_m^j\right]\right)$