PTD rebuttal: additional comparisons

Anonymous submission. Within each figure every column/row starts from the same initial noise. Every section shows the same 18 prompts of the 100-prompt test set, numbered 1 to 18 in the same order (selected examples; the tables cover all 100 prompts). In every table Δ = PTD − other (a positive value favors PTD); * = significant.

1. Curvature head on Wan2.1-1.3B: K=0 vs. K=1

K=1 (learned curve) vs. K=0 (straight-chord queries), trained identically: same prompts, seed, teacher, query budget and updates; only the curve that places the teacher queries differs. Columns: 50-step teacher | K=0 | K=1.

VBench (946 prompts × 5 videos):

totaldynamic degreetemporal flickering
K=0 (straight chord)83.7174.799.21
K=1 (learned curve)83.8476.798.75

Blinded pairwise judge (100 prompts × 3 seeds; Δ = learned curve − K=0):

dynamicstaticnaturalnessdiversity
curve − K=0+.05+.10*+.21*−.00

Reading. K=0 produces red and blue color artifacts that the learned curve does not; the judge rates the curve more natural (+.21) with better per-frame quality (+.10).

Columns: teacher (50 steps) · K=0 (straight chord) · K=1 (learned curve)

prompt

2. Curvature head on MiniMax-H3: K=0 vs. the paper model

K=0 is trained with the paper's H3 recipe (150 updates; same data, teacher, adapters and final-hop march) but without the curvature head, so the two teacher queries per sample lie on the straight chord; it is compared with the paper's model (learned curve). Columns: 50-step teacher | paper model | K=0.

Blinded pairwise judge (100 prompts × 3 seeds; Δ = paper model − K=0):

paper model − K=0dynamicstaticnaturalnessdiversity
all 100 prompts+.09*+.03+.03−.07
50 action prompts+.14*+.01−.01−.12*

Reading. The paper model (learned curve) is more dynamic than K=0 (+.09; +.14 on the action prompts), and K=0 adds fine-scale frame-to-frame flicker beyond the teacher (finest-band power 120% of the teacher for K=0, 105% for the paper model); overall motion is unchanged.

Columns: teacher (50 steps) · paper model (learned curve) · K=0 (no curvature head)

prompt

3. Matched-budget baselines, Wan2.1-14B

Each baseline is trained with PTD's Wan recipe (64 updates, batch 256, the same teacher, 4-step inference) with one component changed, and compared with PTD@64. Rows: PTD, Euler endpoint regression, LMD | PTD, MeanFlow-style, MeanFlow-style with 75% flow-matching samples. Judge Δ = PTD − baseline:

dynamicstaticnaturalnessdiversity
PTD − Euler endpoint regression+.64*+1.04*+.73*+.05
PTD − LMD (two-time map)+.32*+.08*−.11*+.86*
PTD − MeanFlow-style+1.04*+2.00*+1.91*+.69*
PTD − MeanFlow-style, 75% r = t+.73*+1.43*+1.34*+.50*

Top: PTD · Euler endpoint · LMD — Bottom: PTD · MeanFlow-style · MeanFlow-style 75%

prompt

4. DMD on Wan2.1-14B

LightX2V's public four-step Wan2.1-14B student (distribution matching, its own sampler) vs. PTD; seeds 0–2 per row. Judge Δ = PTD − DMD:

dynamicstaticnaturalnessdiversity
PTD − DMD+.11−.07+.40*+.83*

Top: PTD, seeds 0–2 · Bottom: DMD, seeds 0–2

prompt