ZY-Dijkstra commited on
Commit
7b3e075
·
verified ·
1 Parent(s): cf05005

Model card: training line

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -18,7 +18,7 @@ Disen-Forcing is an autoregressive text-to-video model distilled from Wan2.1-T2V
18
 
19
  - `disen_forcing.safetensors`: model weights (bf16).
20
  - Student: Wan2.1-T2V-1.3B, initialized from the Causal Forcing chunk-wise ODE checkpoint; teacher: Wan2.1-T2V-14B.
21
- - Training: DMD distillation with learnable context routing, EMA-based motion regularization and the KV caching gradient (see the [code repository](https://github.com/Dijkstra14/Disen-Forcing)).
22
 
23
  ## Usage
24
 
 
18
 
19
  - `disen_forcing.safetensors`: model weights (bf16).
20
  - Student: Wan2.1-T2V-1.3B, initialized from the Causal Forcing chunk-wise ODE checkpoint; teacher: Wan2.1-T2V-14B.
21
+ - Training: DMD distillation with learnable context routing and EMA-based motion regularization.
22
 
23
  ## Usage
24