- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper distills three capabilities into LoRAs on a base video diffusion model: CFG distillation, four-step sampling, and long-context error correction. These LoRAs are then plugged into downstream fine-tunes of that base without target training. CFG and few-step distillation are kept in separate LoRAs, so guidance strength can be tuned through the CFG LoRA weight. On SCOPE and Wan2.2-Fun-5B-Control, the transferred LoRAs clearly beat naive four-step sampling and come close to per-target distillation. The long-context LoRA improves long rollouts on ReWorld and Matrix-Game 3.0. Transfer to 54 downstream models across three backbone families is shown qualitatively.
Critical strengths
- Practical value and breadth. Repeating distillation for every downstream video model is a real cost. The paper tries its recipe on an unusually wide range of downstream models: full fine-tunes, task LoRAs, ControlNet branches, and models with extra output channels. Where quantitative results are given, the transferred adapters come close to per-target distillation.
- Decoupled guidance control. Distilling CFG separately from few-step sampling lets each downstream task adjust guidance without breaking few-step generation. This simple design makes a single adapter set reusable across tasks with different guidance preferences.
Critical weaknesses
- Limited novelty, and no comparison with closely related work. Distilling acceleration into a LoRA once and plugging it into fine-tunes of the same base was already done for images by LCM-LoRA [1]. Guidance distillation with continuous control over guidance strength already exists: Meng et al. [2] condition the student on , and AGD [3] conditions adapters on the guidance scale and composes them with ControlNet and IP-Adapter. Public four-step, CFG-free distillations also exist for both Wan backbones used in the paper: the lightx2v StepDistill-CfgDistill LoRA for Wan2.1-14B [4], and Wan2.2-TI2V-5B-Turbo [5], whose codebase FlashMotion (listed in Tab. 5) builds on. The long-context LoRA applies LongLive's Streaming Long Tuning [6]. None of these is used as a baseline.
- The 54-model claim is mostly qualitative. Tab. 5 states that each entry "is evaluated" but gives no criterion or metric for success. Apart from the H3 camera-control attenuation discussed in App. C.3, the paper does not say whether any attempted model failed. The demo shows 18 of the 54 models, with one selected example each. Quantitative results cover only two acceleration targets and two long-context targets, all in the Wan2.2-5B family. Wan2.1-14B and MiniMax-H3 have no quantitative results.
- Tab. 2 has no locally run native 40-step result. Tabs. 1/2 do not include the transferred coupled CFG+step LoRA, which is the direct ablation of the decoupling design. is 3 in Tab. 1 and 1 in Tab. 2, and the paper does not say how it was chosen.
- The long-context evaluation does not isolate the long-context LoRA. Tab. 3 has no four-step few-step-only row, even though such outputs appear in Figs. 19–21. The ReWorld group also omits ReWorld's official 4-step DMD LoRA [7, Sec. 2.4], while the Matrix-Game group does include an official distilled baseline. In addition, the long-context LoRA's training setup is not given: which causal AR Wan base, what data, and what rollout length (see Q2).
- The intro claims that the CFG LoRA weight adjusts guidance "in a near-linear manner, as we observe empirically", but the evidence is limited to qualitative examples (Figs. 5–6, 14–18), with no quantitative measurement of the relationship. Fig. 6A and 6B use different coupled checkpoints, so the two panels cannot be compared directly.
[1] Luo et al., LCM-LoRA: A Universal Stable-Diffusion Acceleration Module, arXiv:2311.05556, 2023. [2] Meng et al., On Distillation of Guided Diffusion Models, CVPR 2023. [3] Perez Jensen & Sadat, Efficient Distillation of Classifier-Free Guidance using Adapters (AGD), TMLR 2025. [4] lightx2v, Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v and Wan2.1-T2V-14B-StepDistill-CfgDistill-Lightx2v, Hugging Face model releases, 2025. [5] Wan2.2-TI2V-5B-Turbo, GitHub: quanhaol/Wan2.2-TI2V-5B-Turbo, 2025. [6] Yang et al., LongLive: Real-time Interactive Long Video Generation, ICLR 2026. [7] Chen et al., ReWorld: An Interactive World Model with Long-Horizon Memory, arXiv:2608.23565, 2026.
Questions for authors
Clarifications (answers could change my assessment):
- What exactly does "verified" mean for each of the 54 models in Tab. 5? Were any models tried and then excluded or found to fail?
- Where did the long-context adapter come from, and how was it trained? In the supplementary material, the ReWorld "+Long" video is long-reworld-official.mp4 (curve id official), and the Matrix-Game "+Long" video is long-matrix-reworld.mp4 (curve id reworld). ReWorld's paper describes an official 4-step DMD LoRA trained with streaming self-forcing rollouts. Is the "+Long" adapter in Tab. 3 the one distilled on an AR Wan base as described in Sec. 4.5, or ReWorld's official LoRA, possibly transferred to Matrix-Game? Please give its training details (base model, data, rollout length, iterations, rank), and add the official ReWorld LoRA as a baseline in Tab. 3.
- How was chosen for Tab. 1 (3) and Tab. 2 (1)? Was the test set used? Which adapter variant appears in Figs. 4 and 9, and how does it differ from the one used in the tables?
Additional evidence requested (needed to support claims already made):
- Please add the following under the same protocol:
- a local native 40-step row for Tab. 2;
- the transferred coupled CFG+step LoRA in Tabs. 1/2;
- a four-step few-step-only row in Tab. 3.
- For a subset of the 54 models that includes Wan2.1-14B and H3 targets, please report per-model quantitative results (one quality metric plus one task-fidelity metric, e.g., camera or action following) for native, naive four-step and transferred inference.
- Please give a quantitative curve of text alignment and quality versus , compared with native CFG at , to support the "near-linear" guidance claim.
Suggestions (would not change my rating by themselves):
- Compare against public Wan step/CFG-distilled releases (lightx2v StepDistill-CfgDistill LoRA for Wan2.1-14B, Wan2.2-TI2V-5B-Turbo), against AGD-style -conditioned guidance adapters, and against a short target fine-tune that starts from the transferred LoRA. Report wall-clock latency, and analyze when transfer fails, e.g., as a function of how far the fine-tuned weights have drifted from the base.
8 comments
W3 and Q4 (rows 1–2): No local native 40-step row for Tab. 2; Tabs. 1/2 lack the transferred coupled CFG+step LoRA.
- Native 40-step row for Tab. 2. The native reference in Tab. 2 is the official PAI-Bench-C leaderboard entry, which is already reported for 40-step native inference under the official protocol, so a local rerun would duplicate it. The new Wan2.1-14B and H3 tables include locally run native rows (50-step + CFG 6; 40-step).
- Coupled CFG+step LoRA. Tabs. 1 and 2 report video-quality metrics, whereas a coupled and a separately weighted adapter differ mainly in text alignment, i.e., in how guidance can be controlled. At the default operating point both give similar video quality, so a coupled row in these tables would not reveal the difference, and we did not add it. The relevant ablation is Fig. 6: the coupled LoRA alone does not provide control over text guidance, and scaling it to strengthen guidance degrades video quality, whereas our separately weighted CFG LoRA provides both adjustable guidance and preserved four-step quality.
W4, Q2 and Q4 (row 3): Where does the long-context adapter come from and how was it trained? Add the official ReWorld LoRA and a four-step few-step-only row to Tab. 3.
- Provenance. ReWorld's official 4-step LoRA is our +Long adapter; the file and curve names in the supplement (
official,reworld) refer to this one file. Adding "official ReWorld LoRA" as a separate Tab. 3 row would duplicate +Long. The same file is applied unchanged to Matrix-Game, with no Matrix-Game training. - Training. The long-context LoRA is trained on a frozen causal AR Wan base with LongLive's Streaming Long Tuning configuration and training prompts, for 2,200 iterations at rank 128 (Sec. 3.4 and App. A, marked in blue).
- Few-step-only row. Added to Tab. 3 (blue rows): ReWorld 73.05 vs. 75.77 for +Long (base 73.51); Matrix-Game 83.12 vs. 84.34 (native 83.73). With 3 seeds per scene, +Long is ahead by 1.19 on average, in 9/10 scenes. The long-rollout gain therefore comes from the long-context LoRA.
Q5: Per-model results for Wan2.1-14B and H3 targets (one quality and one task-fidelity metric; native, naive four-step, transferred).
Native / naive four-step / ours:
- Wan2.1-Fun-14B-Control (Tab. 6, 600 cases): Aesthetic / Imaging / DOVER 0.507 / 0.732 / 10.29, 0.439 / 0.669 / 7.46, 0.505 / 0.734 / 9.90 (retrained task-specific distillation 0.512 / 0.710 / 9.78).
Tab. 6 (new App. F.1): Wan2.1-Fun-14B-Control (depth), 600 PAI-Bench-C cases, 480p, 81 frames.
| Method | Aesthetic ↑ | Imaging ↑ | DOVER ↑ | NFE |
|---|---|---|---|---|
| Native 50 step + CFG 6 | 0.507 | 0.732 | 10.29 | 100 |
| Naive 4 step | 0.439 | 0.669 | 7.46 | 4 |
| ControlNet-specific distillation (retrained) | 0.512 | 0.710 | 9.78 | 4 |
| LongLive-Plug (λ_step=1, λ_cfg=1) | 0.505 | 0.734 | 9.90 | 4 |
- H3-Fun-ControlNet-Union (Tab. 7, 600 cases): depth si-RMSE 1.608 / 1.521 / 1.489, F1 0.068 / 0.069 / 0.077, DOVER 9.04 / 9.06 / 9.78.
Tab. 7 (new App. F.1): MiniMax-H3-Fun-ControlNet-Union (depth) and adapter/sampler ablation, 600 PAI-Bench-C cases, 480p, 121 frames.
| Method | SSIM ↑ | F1 ↑ | si-RMSE ↓ | mIoU ↑ | DOVER ↑ | LPIPS ↑ | NFE |
|---|---|---|---|---|---|---|---|
| Native 40 step | 0.615 | 0.068 | 1.608 | 0.556 | 9.04 | – | 40 |
| Naive 4 step | 0.652 | 0.069 | 1.521 | 0.573 | 9.06 | 0.458 | 4 |
| LongLive-Plug adapter + Euler update (ablation) | 0.624 | 0.073 | 1.551 | 0.564 | 9.62 | 0.496 | 4 |
| LongLive-Plug (re-noising sampler) | 0.623 | 0.077 | 1.489 | 0.570 | 9.78 | 0.514 | 4 |
- SolarWM-H3 (Tab. 8, 120 clips, camera following): RotErr 83.8 / 89.3 / 71.1, imaging 69.6 / 62.1 / 69.7.
Tab. 8 (new App. F.2): SolarWM-H3 (camera-controlled world model on MiniMax-H3), 120 official test clips, 158 frames.
| Method | NFE | RotErr ↓ | TransErr ↓ | Trans. ratio (→1) | LPIPS ↓ | Aesthetic ↑ | Imaging ↑ |
|---|---|---|---|---|---|---|---|
| Native (30 points) | 29 | 83.8 | 2.72 | 1.030 | 0.372 | 0.529 | 69.6 |
| Naive 4 step | 4 | 89.3 | 3.10 | 1.022 | 0.394 | 0.494 | 62.1 |
| SolarWM official distillation (Stage 2) | 4/chunk | 107.7 | 3.58 | 0.899 | 0.416 | 0.531 | 71.1 |
| LongLive-Plug (4 step, no SolarWM training) | 4 | 71.1 | 2.53 | 1.035 | 0.347 | 0.540 | 69.7 |
- Wan-Alpha (Wan2.1-14B) (Tab. 9): alpha IoU 0.857 / 0.540 / 0.838.
Tab. 9 (new App. F.3), Wan-Alpha (Wan2.1-14B), RGBA generation, 40 prompts.
| Method | NFE | Matte IoU ↑ | Matte MAD ↓ | CLIP-T ↑ | Imaging ↑ |
|---|---|---|---|---|---|
| Native (50 steps + CFG) | 100 | 0.857 | 3.80 | 26.31 | 0.507 |
| Naive 4-step | 8 | 0.540 | 10.67 | 25.30 | 0.286 |
| LongLive-Plug (transferred) | 4 | 0.838 | 5.29 | 27.13 | 0.548 |
W5 and Q6: Quantitative curve of text alignment and quality vs. λ_cfg against native CFG, to support "near-linear"; Fig. 6A/6B use different coupled checkpoints.
- Guidance sweep (App. F.4, marked in blue). On the full SCOPE CrossFPS test set (1,378 clips), for the CFG LoRA alone and composed with the few-step LoRA, Tab. 10 shows that text alignment increases monotonically with λ_cfg, and compares it with native CFG at the nominally matched scale w = 1 + 4λ_cfg. Native CFG is also monotone; the CFG LoRA follows the same trend with a weaker response (CLIP 19.30 vs. 20.54 at λ_cfg = 1, i.e., w = 5), and λ_cfg = 2 matches the FVD of native CFG 5 (379.9 vs. 382.0) with one pass per step instead of two. We therefore describe λ_cfg as a monotone guidance control and no longer claim it is near-linear (Introduction, marked in blue). In the four-step setting, a large λ_cfg trades distribution fidelity for alignment (FVD 486.8 → 649.6), which we state as a limitation.
Tab. 10 (new App. F.4): CLIP text alignment (↑) vs. λ_cfg on SCOPE, 1,378 CrossFPS test clips. λ_cfg = 0 removes the CFG LoRA; native CFG uses two forward passes per step at the matched scale w = 1 + 4λ_cfg.
| Setting | λ_cfg=0 | 0.5 | 1 | 2 | 3 | 5 |
|---|---|---|---|---|---|---|
| CFG LoRA (30 steps) | 18.03 | 18.55 | 19.30 | 20.37 | 20.91 | – |
| Native CFG (30 steps, w = 1+4λ_cfg) | 18.03 | 19.64 | 20.54 | 21.45 | 21.89 | – |
| Few-step + CFG LoRA (4 steps) | 22.22 | – | 22.38 | 22.50 | 22.56 | 22.57 |
- Fig. 6A/6B. Both panels use the same coupled checkpoint, prompt, seed and inputs; panel B varies only that checkpoint's weight. We revised the caption to "(B) uses different checkpoint weights" (marked in blue).
Q7: Compare with public Wan step/CFG-distilled releases, AGD-style adapters and a short target fine-tune from the transferred LoRA; report wall-clock latency; analyze when transfer fails, e.g., as a function of weight drift from the base.
- lightx2v StepDistill-CfgDistill LoRA: compared in new Fig. 22 (see W1).
- Wall-clock latency. On Wan2.1-Fun-14B-Control (480p, 81 frames, one H100), generating one video takes 411 s with native 50-step + CFG inference (100 NFE) and 19 s with our four-step transfer (4 NFE), a 21× speed-up.
- When transfer fails. Transfer breaks when the adapter is used outside the regime it was distilled in: a different solver (TalkVerse UniPC: Sync-C 4.50 vs. 5.07), long AR rollouts without the long-context LoRA (Tab. 3), or a large CFG-LoRA weight in the four-step composition (FVD 486.8 → 649.6 at λ_cfg = 5).
We thank the reviewer for the precise questions. We answer each Weakness together with the Questions that refer to it.