sd15-flow-matching

# Pseudocode - implementation details TBD
x_t = noise
for t in reversed(timesteps):
    v = student_unet(x_t, t, text_embeddings)
    x_t = step(x_t, v, t)  # v-prediction update
image = vae.decode(x_t)

v* = α · ε - σ · x₀  (target)
v̂ = student(x_t, t)  (prediction)
L_flow = MSE(v̂, v*)

L_kd = 1 - cosine_sim(
    student_features.mean(spatial), 
    teacher_features.mean(spatial)
)