通过条件掩码扩散语言模型实现嵌入反向
Embedding Inversion via Conditional Masked Diffusion Language Models
AI总结：
本文提出通过条件掩码扩散语言模型实现嵌入反向，无需访问目标编码器，通过并行去噪技术高效恢复标记。
摘要：
我们将嵌入反向视为条件掩码扩散，通过迭代去噪而非顺序自回归生成来并行恢复所有标记。通过自适应层归一化将目标嵌入作为条件输入给掩码扩散语言模型，仅需8次前向传递，推理时无需访问目标编码器。在三个嵌入模型上，针对32个标记序列，通过并行去噪实现标记恢复，无需要求编码器访问、迭代修正或架构特定对齐。源代码和实时演示可在https://github.com/jina-ai/embedding-inversion-demo获取。
英文摘要：
We frame embedding inversion as conditional masked diffusion, recovering all tokens in parallel through iterative denoising rather than sequential autoregressive generation. A masked diffusion language model is conditioned on the target embedding via adaptive layer normalization, requiring only 8 forward passes with no access to the target encoder at inference time. On 32-token sequences across three embedding models, the method achieves token recovery through parallel denoising without requiring encoder access, iterative correction, or architecture-specific alignment. Source code and live demo are available at https://github.com/jina-ai/embedding-inversion-demo.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
