Unverified90% confidenceFactTime unknown
Mercury 2 将扩散技术（Diffusion）应用到语言生成领域，通过并行去噪过程同步优化多个 Token，而非逐一自回归生成
1
Sources
90%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
Unverified掩码扩散框架将加噪重新定义为token掩码替换，使扩散过程在离散文本域得以自洽57% similarUnverifiedThe text encoder is responsible for converting natural language prompts into high-dimensional embedding vectors for use by the diffusion model's denoising process55% similarVerifiedWhisper's architecture adopts an encoder-decoder structure, where the encoder transforms the mel spectrogram of audio into high-dimensional feature representations, and the decoder generates text output word by word in an autoregressive manner.55% similarUnverifiedMidjourney uses CLIP's text encoder to convert prompts into conditioning vectors that guide the denoising process of the diffusion model.53% similarUnverified现代歌声合成方法转向以扩散模型或变分自编码器为骨干的生成架构，早期方法依赖规则驱动的参数合成器如VOCALOID53% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/14034API
curl https://kongchang.com/api/v1/knowledge/claims/14034MCP
get_claim(id=14034)