Unverified50% confidenceFactExact time
YuE2 underlyingly uses a 28-layer Mixture-of-Transformers (MoT) architecture, running autoregressive (AR) and non-autoregressive (NAR) expert types in parallel, which share attention computation but use separate normalization, projection, and feed-forward layers.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWan 2.2 属于扩散变换器（Diffusion Transformer，DiT）架构66% similarUnverified在Transformer架构中，每个层由多头自注意力机制和多层感知机（MLP）两个核心子模块组成65% similarUnverifiedToday's mainstream visual encoders are highly similar at the underlying architecture level, with almost all being based on Transformers.63% similarUnverifiedLaya pairs a bidirectional ModernBERT-large encoder with a Transformer head trained from scratch that scores [MASK] option tokens.62% similarUnverifiedKrea 2模型在底层架构上融合了扩散Transformer（DiT）思路，优于早期的U-Net架构模型62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1029277API
curl https://kongchang.com/api/v1/knowledge/claims/1029277MCP
get_claim(id=1029277)