- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes a training-free acceleration method for diffusion-based world models. It combines adaptive anchor refresh with geometry-aware decomposition of temporally predicted state displacements into shared and residual components. Separate gains for these components are estimated online from completed full-anchor transitions using closed-form regularized regression, without additional transformer evaluations. Experiments report 4.25× acceleration on Voyager and 1.96× on Aether generation, with improved fidelity relative to WorldCache.
Critical strengths
The author provides a useful geometric extension to temporal state prediction. The method uses geometric correspondence to define shared and residual displacement components while retaining token-specific variation. Online calibration is computationally lightweight and uses only completed anchor transitions. Meanwhile, the controlled experiments show that geometry contributes beyond generic calibration. Table 4 compares operators under the same full-evaluation schedule while allowing each variant to evolve its own sampler trajectory. GeoTrans improves PSNR by 0.62 dB over global scalar calibration and 0.82 dB over shuffled correspondence.
Critical weaknesses
-
The cross-model geometry construction is insufficiently specified. Appendix A.2 clearly describes supplied geometry for Voyager, but Appendix D.3 does not establish equally clearly how the fixed correspondence map is obtained for each Aether task. In particular, the reconstruction setting requires clarification of which depth and camera quantities are available before inference and which are prediction targets.
-
The reliability of the incremental quality improvements is not quantified. Although the full-benchmark evaluation is a strength, several improvements over WorldCache are relatively small: +0.33 dB PSNR and +0.23/+0.16 WorldScore points on Voyager, and +0.17 dB PSNR on Aether. The paper does not report paired scene-level uncertainty or variability across generation seeds. Consequently, it is difficult to assess how consistently the quality advantage holds, particularly beyond the Voyager operator ablation.
Questions for authors
-
For Aether generation and reconstruction, what exact depth and camera information is used to construct the fixed correspondence map, and when is it available?
-
For LingBot, how are the correspondence map and gain updates constructed under the four-call chunk schedule?
-
Were accelerated methods and full-step references evaluated using matched random seeds?
-
Can you provide paired scene-level confidence intervals for the main quality differences, especially Voyager WorldScore and Aether PSNR, and state how many generation seeds were used?
5 comments
Response to Reviewer pLw6 (Part 2)
R3-Q1: Depth and camera information used for Aether correspondence maps and its availability.
Response to R3-Q1: Generation and video-depth reconstruction use fixed correspondence maps, while camera-pose reconstruction uses analytic raymap constraints.
Table 3. Geometry sources and availability for Aether tasks.
| Task / protocol | Depth | Camera information | Availability and cost |
|---|---|---|---|
| Generation | MoGe depth from input image | Supplied trajectory (raymap) and intrinsics | Before denoising; 0.9 s/scene included in runtime |
| Reconstruction: video-depth | MoGe depth from observed frames | MoGe intrinsics; optical flow and RANSAC-PnP poses using MoGe 3D points | Before each window; all preprocessing costs included |
| Reconstruction: camera-pose | Not used | Analytic camera-ray constraints; no external pose estimate | Within the timed sampler |
Correspondences use integer pixel centers, visibility checks, and a 10% relative depth-consistency threshold. Projected patches are grouped by reference-grid cells across latent frames. Maps remain fixed within each sampling window or rollout. Video-depth falls back to scalar calibration when retained groups cover less than 10% of tokens.
For camera-pose, skipped-step raymaps are constrained to affine ray directions and a shared ray origin per frame. Evaluated depth and poses remain Aether's predictions; no evaluation annotations, full-step references, or denoising outputs are used to construct maps.
Revisions. Appendix D.3 now details task-specific geometry sources, construction, availability, and runtime accounting.
R3-Q2: Correspondence map construction and gain updates under LingBot’s four-call chunk schedule.
Response to R3-Q2: The four-call schedule separates first-chunk calibration from subsequent prediction:
- Geometry construction: Before denoising, MoGe depth and supplied camera poses/intrinsics define the correspondence map. Poses are interpolated to latent frames, and intrinsics are scaled to 464×832. Correspondences index the 29×52 token grid of the intermediate DiT block-stack state before the prediction head. The map remains fixed throughout the rollout.
- First-chunk calibration: All four calls execute exactly. The first three states () form a proposal for the fourth before its exact output is observed. Shared/residual components are separated by the chunk-local projector, and the fourth exact state provides the sole calibration target for fitting two gains (Eq. 20).
- Subsequent chunks: The first three calls remain exact; the fourth block-stack state is predicted using fixed gains. No further calibration occurs. Prediction history and grouping remain chunk-local, while the native prediction head and KV updates always execute fully.
Across four chunks, both methods replace three block-stack evaluations: 3/20 = 15% including KV updates, or 3/16 = 18.75% of denoising calls. These represent call-count reductions.
We evaluate per-chunk fidelity against matched full rollouts over the first 49 of 61 generated frames.
Table 4. Per-chunk PSNR with first-chunk-only calibration.
| Chunk | Frames | WorldCache ↑ | GeoTrans ↑ (gain [95% CI]) |
|---|---|---|---|
| 0 (calibration) | 1–13 | 44.38 | 44.39 (+0.00 [−0.01, 0.01]) |
| 1 | 14–29 | 13.33 | 15.05 (+1.72 [1.57, 1.89]) |
| 2 | 30–45 | 11.90 | 13.75 (+1.85 [1.36, 2.38]) |
| 3 | 46–49 | 10.80 | 12.32 (+1.51 [0.99, 2.09]) |
Gains denote GeoTrans − WorldCache, with paired scene-level 95% CIs. GeoTrans maintains positive PSNR gains across all three prediction chunks.
Revisions. Appendix D.3 now clarifies geometry construction, the pre-head interface, calibration ordering, and fixed-gain reuse. Table 2 specifies the call-count denominator.
Response to Reviewer pLw6 (Part 3)
R3-Q3: Matched random seeds between accelerated methods and full-step references.
Response to R3-Q3: All full-step and accelerated evaluations use matched random seeds, initial diffusion noise, and conditioning-latent sampling. Fidelity is measured against the corresponding full-step output for each scene–seed pair.
The seed configurations are:
- Voyager: Seed 0 for baseline reproduction (40 single-scene and 10 three-scene prompts); seeds 0, 1, 2 for variability analysis.
- Aether: Seed 42 for generation; seeds 42, 43 for reconstruction.
- LingBot: Seed 42 for chunked generation.
For closed-loop rollouts, each method conditions subsequent scenes on its own generated final frame, preserving autoregressive error propagation.
Revisions. Appendices D.2–D.3 now document seed settings and conditioning-latent sampling, while Section 5.1 clarifies the pairing protocol.
R3-Q4: Paired scene-level confidence intervals and generation seed counts for Voyager WorldScore and Aether PSNR.
Response to R3-Q4: We provide 95% paired bootstrap CIs using 10,000 scene-level resamples. For Voyager's three-seed evaluation, differences are averaged across seeds per scene before resampling. The original benchmark generation results use one seed per prompt; revised Aether reconstruction results average two seeds.
Table 6. Voyager WorldScore-Static at matched denoising cost.
| Evaluation | Scenes × Seeds | WorldCache | GeoTrans | Δ [95% CI] |
|---|---|---|---|---|
| Seeds 0, 1, 2 | 20 × 3 | 62.43 | 62.24 | −0.19 [−1.26, 0.84] |
| Seed 0 | 40 × 1 | 64.42 | 65.21 | +0.79 [−0.56, 2.17] |
Δ denotes GeoTrans − WorldCache. Voyager PSNR/LPIPS CIs are provided in Table 1 of Response to R3-W2.
Table 7. Aether generation: 20 scenes, seed 42.
| Metric | WorldCache | GeoTrans | Δ [95% CI] |
|---|---|---|---|
| PSNR ↑ | 34.15 | 34.97 | +0.81 [0.54, 1.10] |
| LPIPS ↓ | 0.0275 | 0.0192 | −0.0083 [−0.0095, −0.0072] |
| SSIM ↑ | 0.9681 | 0.9747 | +0.0067 [0.0052, 0.0082] |
Both methods share identical inputs, noise, warm-up (10 steps), and FULL-evaluation positions (30/50 steps). Two-seed Aether reconstruction results appear in Table 2 of Response to R3-W2.
Results. Voyager WorldScore CIs include zero in both panels, while PSNR/LPIPS improvements are consistent across three seeds. Aether generation also shows positive fidelity improvements with CIs excluding zero. We have revised the discussion to distinguish these statistically supported fidelity gains from the inconclusive WorldScore differences.
Revisions. The experimental appendix now documents paired resampling, seed counts, and confidence intervals for each evaluation.
We sincerely thank the reviewer for the valuable feedback, which has helped us improve the clarity and rigor of our paper. We hope our responses, additional experiments, and revisions have adequately addressed your concerns. We remain active throughout the rebuttal period and welcome any further questions or discussion. If our clarifications have resolved your concerns, we would greatly appreciate your consideration of a more favorable assessment.
Response to Reviewer pLw6 (Part 1)
We thank the reviewer for their valuable comments and constructive feedback. We address each point below.
R3-W1: Cross-model geometry construction and availability of depth and camera information for Aether tasks.
Response to R3-W1:
No evaluation annotations, full-step references, or Aether denoising outputs are used for map construction.
Revisions. Appendix D.3 now clarifies geometry sources, availability, prediction targets, and runtime accounting for each task. Fixed correspondence maps apply to generation and video-depth; the scalar fallback below 10% correspondence support applies only to video-depth, while camera-pose uses analytic raymap constraints.
R3-W2: Statistical reliability of quality improvements over WorldCache across scenes and generation seeds.
Response to R3-W2:
For Voyager, we evaluate 20 paired scenes across seeds 0, 1, and 2, using identical inputs and initial noise. Both methods use 12 FULL evaluations, with matched denoising times (240.7 s WorldCache vs. 240.4 s GeoTrans). Unlike Table 1’s default WorldCache setting (3.65× vs. 4.25×), this comparison isolates fidelity at equal inference cost.
Table 1. Voyager quality differences (GeoTrans − WorldCache) with paired 95% CIs.
PSNR and LPIPS improvements are consistent across seeds, with CIs excluding zero, while WorldScore differences remain statistically inconclusive.
For Aether generation, 20 paired scenes with identical warm-up and FULL-evaluation positions yield +0.81 dB PSNR (95% CI [0.54, 1.10]).
For Aether reconstruction, we evaluate two seeds on MPI-Sintel using WorldCache’s 30-step protocol at comparable inference cost.
Table 2. Aether reconstruction results by seed with paired 95% CIs.
GeoTrans reduces AbsRel by 14.6%/14.5% and ATE by 9.6%/17.3% for seeds 42/43. Paired 95% CIs for GeoTrans − WorldCache are [−0.087, −0.038]/[−0.083, −0.034] (AbsRel) and [−0.037, −0.003]/[−0.095, −0.008] (ATE). All intervals exclude zero. Speedups are relative to 30-step Aether.
Revisions. We added per-seed results, paired confidence intervals, and resampling procedures to the experimental appendix. The revised discussion distinguishes robust fidelity gains from smaller, statistically inconclusive WorldScore differences.