- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This work proposes KalmanCache+, a training-free framework for accelerating diffusion transformers by selectively skipping transformer-stack evaluations. A channel-wise Kalman filter estimates residual dynamics and predictive covariance, which determines whether to perform a full evaluation or reconstruct the output from cached residuals. Skipped computations use a blend of residual reuse, linear extrapolation, and a stored extrapolation mismatch. Full evaluations update both the estimator and the correction memory.
Critical strengths
The method couples the decision of when to recompute with the reconstruction of skipped outputs. Using full evaluations to improve subsequent decisions and reconstructions is a reasonable approach to managing caching errors.
The paper includes additional seeds, schedule-replay experiments, compute-budget targeting, sensitivity analysis, and overhead measurements. The reported 2.12% sampling-loop overhead over plain reuse under the same DiT schedule supports the practicality of the implementation.
Critical weaknesses
Appendix C.2 provides useful correlations with plain-reuse error on fully computed DiT trajectories, but does not directly validate the error of the actual reconstructed residual on trajectories affected by caching. The central concern is whether the gate reliably avoids large-error SKIP decisions.
Table 4 compares extrapolation with a reconstruction that includes both uncertainty-dependent blending and the stored mismatch. Although Appendix E.4 investigates blending, a controlled ablation removing only the C-DOB term while preserving the schedule and weights is still needed.
What advantages does KalmanCache+’s online adaptation offer over ERTACache’s offline-calibrated caching and error correction [1]? ERTACache explicitly analyzes cache-induced error accumulation and introduces timestep adjustment and residual error rectification.
How does C-DOB differ from FoCa’s predictor–corrector approach [2], and does it provide additional benefits under matched FULL/SKIP schedules? FoCa already combines feature forecasting with a correction mechanism to improve stability under aggressive skipping. Could the authors clarify the distinction between C-DOB and the predictor–corrector approach in FoCa?
[1] Peng, Xurui, et al. "Ertacache: Error rectification and timesteps adjustment for efficient diffusion." International Conference on Learning Representations. Vol. 2026. 2026.
[2] Zheng, Shikang, et al. "Forecast then calibrate: Feature caching as ode for efficient diffusion transformers." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 40. No. 16. 2026.
Questions for authors
Can you evaluate the relationship between the uncertainty score and the actual reconstructed-residual error on caching-generated trajectories across all three backbones? In particular, how frequently do large errors occur below the SKIP threshold, and does the score remain informative after controlling for skip length?
Can the key FLUX comparisons be conducted using a common implementation and full-computation reference? Please also report speedups relative to each implementation’s own baseline.
Can the authors provide matched-latency comparisons with ERTACache [1] and FoCa [2] under consistent implementations and evaluation protocols? Reporting both paired fidelity and reference-free generation quality, together with warmup and calibration costs, would clarify the practical advantage of KalmanCache+.
3 comments
Response to Reviewer 3vjR [2/3]: how does KalmanCache+ compare with ERTACache at near-matched latency?
Response W3 and Q3
ERTACache at near-matched latency
We compare the complete methods near 2× with ERTACache's official code; calibration and warmup-selection costs are in Table R5 ([3/3]).
Without trajectory dumps or fitted coefficients, KalmanCache+ is closer to full computation than ERTACache near 2× on Wan2.1, against its shipped skip list and a stronger list we searched with its procedure. We ran its official code (timestep adjustment and rectification on, coefficients fitted on 10 prompts outside the evaluation set) with our 50 prompts (seed 2; also used by the paper's warmup sweep), reference, and evaluator. On Wan2.1 our gate acts as an adaptive refresh interval, so this compares schedule plus reconstruction with calibrated schedule plus rectification.
Table R4: Wan2.1-1.3B, 81 frames, 832×480, one A100; full computation 196 s. Bold: best per latency group.
| Method | Computed steps | Time (speedup) | PSNR ↑ | SSIM ↑ | LPIPS ↓ |
|---|---|---|---|---|---|
| ERTACache, shipped list | 24 | 95.2 s (2.06×) | 25.65 | 0.8670 | 0.0824 |
| ERTACache, our list (thr. 0.12) | 24 | 95.2 s (2.06×) | 28.24 | 0.9103 | 0.0541 |
| KalmanCache+ (λ=22, W=15) | 22.6 | 90.7 s (2.16×) | 30.11 | 0.9311 | 0.0389 |
| KalmanCache+ (λ=25, W=18) | 25.0 | 99.4 s (1.97×) | 31.61 | 0.9441 | 0.0311 |
| ERTACache, our list (thr. 0.175) | 18 | 73.5 s (2.67×) | 24.18 | 0.8405 | 0.1049 |
| KalmanCache+ (λ=40, W=11) | 18 | 74.3 s (2.64×) | 25.61 | 0.8592 | 0.1010 |
| ERTACache, our list (thr. 0.2) | 16 | 66.0 s (2.97×) | 22.17 | 0.7922 | 0.1485 |
| KalmanCache+ (λ=32, W=7) | 15 | 63.8 s (3.07×) | 21.17 | 0.7553 | 0.1852 |
- Near 2×, from both sides of its latency: the paper's two settings are +4.5 to +6.0 dB closer than its shipped list. Its threshold search with our list-merging rule (its code gives none), run at 24 steps, beats the shipped list by +2.59 dB [1.88, 3.34]; against that list our margins are ([1.43, 2.32], [3.05, 3.73]), with lower LPIPS. On 50 new prompts, settings frozen: +4.23 / +5.39 dB over the shipped list and +1.25 / +2.42 dB over our list ([0.55, 1.82], [1.46, 3.14]; SSIM, LPIPS n.s.).
- Higher acceleration (same procedure, 5 calibration prompts, step counts fixed in advance): at 2.64× vs. 2.67× we are +1.42 dB [0.50, 2.17] while 0.8 s slower (SSIM, LPIPS n.s.), and +0.27 dB with worse SSIM/LPIPS (all n.s.) when ERTACache gets 2.8 s more (2.54×). At about 3× ERTACache is 1.00 dB [0.37, 1.69] ahead with better SSIM and LPIPS, and still 1.09 dB [0.43, 1.75] ahead when we add one FULL step to match its time (66.4 vs. 66.0 s). New prompts: +1.36 / −1.05 dB at the 2.64× / 3× pairs, −0.72 at matched time (2.8 s bracket not re-run).
- FLUX.1-dev: our fastest diffusers point (λ=100, 5.87 s) shows no significant PSNR, LPIPS, or ImageReward difference from ERTACache's released configuration (ImageReward 0.949 vs. 0.902; full computation 1.029), with lower SSIM. ERTACache (8 of 28 steps computed) is 1.5 s faster at 4.37 s (uncached 28 steps: 19.56 dB); we have no point at its speed.
Response to Reviewer 3vjR [3/3]: how does C-DOB differ from FoCa, and what does calibration cost?
Response W4
C-DOB vs. FoCa's predictor–corrector
Both correct forecasts with full computations; with the same schedule and blend they are close, and on identical FULL lists our margin over FoCa grows as FULL steps become scarcer. FoCa applies a BDF2 predictor and a Heun-type corrector per cached module with fixed weights. C-DOB stores one tensor, the extrapolation error of the block-stack residual at a FULL step, and adds it through the next SKIP segment with the trust weight ; with equally spaced FULL steps, is the second backward difference of the last three FULL residuals.
Table R6: FLUX.1-dev, DrawBench-200, the 13/50-FULL schedule of Table R3; block-stack residual only.
| Correction | PSNR ↑ | SSIM ↑ | LPIPS ↓ |
|---|---|---|---|
| C-DOB in our trust-weighted blend | 20.85 | 0.7935 | 0.2160 |
| FoCa-style corrector in the blend | 20.68 | 0.7960 | 0.2156 |
| FoCa-style corrector, full weight | 20.25 | 0.7889 | 0.2216 |
| C-DOB, full weight | 18.42 | 0.684 | 0.376 |
- Same schedule: in the blend, C-DOB is +0.16 dB PSNR and the FoCa-style corrector +0.0025 SSIM, LPIPS within 0.0004, and combining them does not help (20.77 dB). At full weight both fall below the blend, so on FLUX.1-dev the small blend weight matters more than the corrector.
- Complete methods (our re-implementation; its corrector variant fixed by a pre-set rule on 30 of the 200 prompts): on FoCa's uniform schedules (Table R0) KalmanCache+ is above it at 2.6–2.8, 3.8, and 5.0× FLOPs and comparable FULL counts, ImageReward n.s. On our gate's identical FULL lists, KalmanCache+ − FoCa is +0.76 [0.62, 0.89] and +1.63 [1.47, 1.79] dB, LPIPS −0.008 and −0.040 (ImageReward n.s.), at 13.2 and 10.1 FULL steps; at 17.9 it is −0.09 dB [−0.40, +0.22] with worse LPIPS (+0.009 [0.004, 0.015]) and ImageReward (−0.019 [−0.038, −0.003]).
- Scope: our FoCa is within 0.07 dB of TaylorSeer O2, with a smaller ImageReward gain than FoCa reports, so these rows are our best reproduction, not a verdict on FoCa; its times, slower than TaylorSeer's, do not represent FoCa's own latency, so we compare at comparable FULL counts and transformer FLOPs.
Response Q3
Warmup and calibration costs
KalmanCache+ stores no trajectory dumps or coefficients; λ comes from a sweep or a compute-log pilot, the video warmup from full computation.
Table R5: what each method needs before a run.
| KalmanCache+ | ERTACache | FoCa | |
|---|---|---|---|
| Calibration data | compute logs for λ; video warmup vs. full computation | cached and full trajectories of calibration prompts | none |
| Stored per model and step count | settings λ, W only; no fitted tables | skip list, per-timestep coefficients | interval N |
| Offline cost | λ pilot about 0.8 (DiT) and 16 (FLUX) A100-minutes; Wan W sweep: 5 × 50 videos, about 383 A100-minutes, plus 50 references (163) | 50.5 A100-minutes on Wan2.1 (8.5 GB of dumps, then the fit), 4.4 A100-minutes on FLUX.1-dev; list searches excluded | interval N per model |
Warmup steps are FULL steps, included in all FLOPs and latencies.
Planned changes for the revised paper
- Table R1 in Appendix C.4, the Wan2.1 scope in the Limitations, and C-DOB-only rows in Table 4.
- ERTACache and FoCa in Section 2.2, Lim.2 revised; Tables R4–R6 in Appendix G.
- Table 2 split by implementation, Table R0 in Section 4.2, Algorithm 1 and Table 5 moved up to Section 3.
If a comparison is still missing, please tell us; we will prioritize the requested comparisons during the discussion period.
Again, we sincerely thank the reviewer for the constructive feedback. If you feel your concerns have been addressed, we would be grateful if you could reconsider the rating.
Response to Reviewer 3vjR [1/3]: does the gate avoid large-error SKIPs, and what does C-DOB alone add?
We thank the reviewer for the careful reading and valuable feedback. Points shared with other reviewers are in the general response [1/4]–[4/4] and the additional results [1/2]–[2/2], posted at the top of this page; "reply to <reviewer>" means our comments under that reviewer's review.
Response W1/Q1
The score against the actual reconstructed error
Your central concern was whether the gate avoids large-error SKIPs. On both image backbones large errors are rare where it skips, and the score ranks the error beyond skip length; on Wan2.1 the gate is an adaptive refresh interval. Protocol and Table R1: general response [2/4].
Response W2
Removing only C-DOB, with the schedule and weight rule fixed
Table R3: Same FULL/SKIP schedule and weight rule; only the C-DOB term removed. DiT-XL/2: FID-50K at 21/50 FULL steps (the gate's schedule replayed). FLUX.1-dev: DrawBench-200 at 13/50. Wan2.1-1.3B: 50 prompts at 22/50.
† Both seed-1 values are from the rebuttal server, which reproduces the paper's 2.195 (seed 1, replayed) and 2.241 (seed 0, adaptive).
On the calibrated 13-, 20-, and 10-FULL FLUX.1-dev schedules, C-DOB adds +0.20 / +0.09 / +0.04 dB on DrawBench-200 and +0.34 / +0.09 / +0.05 dB on 200 held-out COCO captions, every interval above zero.
Every question, in one line