- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes VLA-Rolling, an inference framework that reuses the unexecuted action tail from the previous prediction round to accelerate flow-matching VLA models. A lightweight confidence estimator evaluates the inherited actions using updated observations and proprioceptive states. The predicted confidence controls the mixture between inherited actions and Gaussian noise and the number of subsequent denoising steps, while the underlying VLA remains frozen. Experiments across four VLA backbones and four simulation benchmarks report reduced action-chunk inference latency and success-rate improvements in most combinations. The paper also includes detailed ablations, analyses, and real-world manipulation experiments.
Critical strengths
- Broad experimental coverage. The method is evaluated across four VLA backbones and four simulation benchmarks, with additional real-world experiments. This breadth provides useful evidence that the proposed inference framework is applicable across different models and environments.
- Promising accuracy–efficiency trade-off. The reported results show reduced action-chunk inference latency across all evaluated model–benchmark pairs, while maintaining or improving task success in most settings. The diagnostic experiments also suggest that inherited action tails can achieve comparable action accuracy with fewer denoising steps.
Critical weaknesses
- Missing comparisons with closely related warm-start methods. The evaluation primarily compares VLA-Rolling against the unmodified backbones and its own ablations. However, closely related methods such as STEP (ICML 2026)(https://arxiv.org/abs/2602.08245) also construct informative warm-start initializations from historical actions and current observations to reduce denoising iterations. Without a direct comparison or a detailed explanation of the methodological differences, it is difficult to establish the incremental benefit of confidence-guided tail reuse. An appropriately adapted comparison under a shared backbone, evaluation protocol, and computational budget would strengthen the contribution.
- The confidence-guided refinement rules are insufficiently justified. The proposed confidence is derived from quantile-normalized log demonstration error, but it is subsequently used both as the action–noise interpolation coefficient and as the basis for allocating denoising steps. In a standard linear flow-matching path, the coefficients are determined by the flow time . The paper does not establish why its learned confidence should play an equivalent role. The specific confidence-to-noise and confidence-to-budget mappings therefore require stronger justification or targeted empirical validation.
Questions for authors
- Comparison under identical denoising budgets. One-step and few-step distillation methods demonstrate that diffusion/flow-based policies can operate with substantially fewer iterations, although they involve additional training. This raises the question of how much of the reported advantage depends on using a 10-step baseline. Could the authors compare Gaussian initialization and VLA-Rolling under identical small, fixed denoising budgets—for example, —while holding other parameters constant?
- Justification of confidence-guided refinement. Could the authors justify using action confidence, supervised by quantile-normalized log action error, to control both confidence-guided renoising (Equation 9) and adaptive-step denoising (Equation 10)? How does this confidence relate to the flow time used by the frozen VLA? Please also clarify how per-action renoising coefficients are reconciled with a shared starting flow time, and provide theoretical justification or targeted ablations supporting these mappings.
- Relationship to STEP. STEP already explores warm-start action generation using historical actions and current observations, followed by a small number of denoising steps. Could the authors provide a detailed distinction between STEP and VLA-Rolling, particularly regarding initialization construction, auxiliary supervision, noise injection, and iteration allocation? A direct comparison, with appropriate adaptation to the same policy backbone, would help establish the benefit of the proposed confidence-guided mechanism.
5 comments
Thanks & Initial Response to Reviewer 1gQK (W2)
[W2 - Insufficient justification for confidence-guided renoising and adapting denoising step rules.]
We thank the reviewer for raising this insightful question. We address the rationale behind from two perspectives.
1. Why use quantile-normalized log demonstration error?
Our confidence estimator is designed to predict the reuse reliability of inherited actions, rather than directly regress the optimal number of denoising steps. We adopt a linear mapping from predicted confidence to the refinement budget, motivated by an early experiment we have conducted during the method design stage.
| Denoising Steps | + LIBERO | + MetaWorld | SmolVLA + LIBERO | SmolVLA + MetaWorld |
|---|---|---|---|---|
| 0 | 3.53 | 2.40 | 3.53 | 2.11 |
| 1 | 3.18 | 2.17 | 3.21 | 1.91 |
| 2 | 2.83 | 1.93 | 2.89 | 1.72 |
| 3 | 2.48 | 1.69 | 2.58 | 1.53 |
| 4 | 2.13 | 1.46 | 2.26 | 1.34 |
| 5 | 1.79 | 1.22 | 1.95 | 1.16 |
| 6 | 1.44 | 0.99 | 1.65 | 0.98 |
| 7 | 1.11 | 0.75 | 1.36 | 0.80 |
| 8 | 0.78 | 0.52 | 1.09 | 0.63 |
| 9 | 0.47 | 0.30 | 0.86 | 0.48 |
| 10 | 0.25 | 0.10 | 0.74 | 0.41 |
Across the evaluated models and benchmarks, action error decreases approximately linearly as the number of denoising steps increases, as shown in Figure 7. This observation provides empirical motivation for our linear confidence-to-budget mapping in Equantion (7) and Equation (10).
To further examine how closely the predicted refinement budget matches the optimal budget defined in Equation (2), we sampled 4,000 demonstrated actions from the LIBERO offline dataset and used demonstration replay to approximate the closed-loop execution context.
| Step N | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | Avg. |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Ground Truth | 2.46% | 9.17% | 15.66% | 14.09% | 12.53% | 11.19% | 7.16% | 4.47% | 4.03% | 19.24% | 5.685 |
| Predicted Step | 0.45% | 9.62% | 12.30% | 14.54% | 13.65% | 8.28% | 8.05% | 4.92% | 7.38% | 20.81% | 6.029 |
The distributions of predicted and oracle budgets exhibit closely matched average step counts and a strong positive spearman correlation (). These results provide additional empirical support for the effectiveness of our budget allocation mechanism. We will incorporate these analyses into the Appendix.
2. How are per-action renoising coefficients coordinated with a shared flow time?
In Equation (9), per-action confidence controls the proportion of inherited action information retained during initialization, rather than assigning an independent flow time to each action position. Equation (10) selects the refinement budget using the minimum confidence among inherited actions.
This choice is supported by our existing ablation in Appendix D.2, Table 6, where minimum-confidence-based allocation achieves lower cumulative trajectory errors than using the mean or maximum confidence on both LIBERO and MetaWorld.
Using the minimum confidence allows the refinement budget to account for the least reliable inherited action, reducing the risk of prematurely terminating denoising based on the average reliability of the chunk. This constitutes a conservative chunk-level scheduling strategy.
Thus, per-action confidence governs the initialization, while the minimum confidence determines the shared refinement schedule. This is a practical approximation design.
Thanks & Initial Response to Reviewer 1gQK (Q1, Q2, Q3)
[Q1 - Comparison under identical denoising budgets.]
To address the reviewer's concern, we conducted additional comparisons using with smaller denoising budgets of 2 and 4 steps, under identical evaluation settings. We did not include the 1-step setting, as VLA-Rolling cannot further reduce the denoising budget in this case. For these experiments, we retained the VLA-Rolling estimator trained under the original 10-step configuration and adjusted its maximum denoising budget accordingly.
| Denoising Budget | Method | Success Rate ↑ | Avg. Denoising Steps ↓ | Inference Time (ms/chunk) ↓ |
|---|---|---|---|---|
| 2 steps | Gaussian Initialization | 97.75% (391/400) | 2.000 | 121.08 |
| 2 steps | VLA-Rolling | 98.00% (392/400) | 1.694 | 118.85 |
| 4 steps | Gaussian Initialization | 96.75% (387/400) | 4.000 | 140.43 |
| 4 steps | VLA-Rolling | 97.50% (390/400) | 3.078 | 132.17 |
The results show that VLA-Rolling maintains success-rate while further reducing denoising steps and inference latency, even under substantially smaller denoising budgets. This suggests that our method's benefits are not limited to the original 10-step baseline, as it continues to reduce redundant computation without sacrificing task performance.
We appreciate the reviewer's suggestion and would be happy to evaluate our method on additional one-step or few-step distilled baselines of interest.
[Q2 - Justification of confidence-guided refinement.]
As detailed in our response to W2, we clarify that confidence serves as an empirical proxy for denoising progress rather than an exact equivalent of flow time. We provide additional evidence supporting the confidence-to-budget mapping, including Spearman correlation coefficients. We also explain that the confidence level for each action renoising during initialization, while the minimum confidence level determines the shared optimization plan, a point supported by the ablation results in Table 6 of Appendix D.2.
[Q3 - Methodological differences between STEP and VLA-Rolling.]
We thank the reviewer for encouraging a clearer comparison with STEP, which have already been included in our related work. Their mechanisms differ in four aspects:
- Initialization Construction: STEP learns to predict a new action sequence from historical actions and current observations, whereas VLA-Rolling directly reuses the unexecuted action tail from the previous inference round.
- Auxiliary Supervision: STEP trains its predictor using action prediction error, while VLA-Rolling learns action-wise reuse confidence from quantile-normalized log demonstration error.
- Noise Injection: STEP employs velocity-aware perturbation to address execution stagnation, whereas VLA-Rolling uses confidence-guided renoising to adjust how much historical action information is retained at each position.
- Iteration Allocation: STEP uses a predefined few-step refinement budget, while VLA-Rolling adaptively determines the remaining denoising steps based on the predicted confidence.
For direct empirical evidence, please refer to our response to W1, where we adapt STEP to the same backbone and compare both methods under a two-step denoising budget.
Thanks & Initial Response to Reviewer 1gQK (W1)
[W1 - Missing comparisons with closely related warm-start methods.]We thank the reviewer for highlighting additional relevant work, which helps us better clarify the scope and positioning of our method. STEP directly predicts a warm-start initialization for flow matching, reducing the subsequent denoising process to two steps. This shares a similar warm-start motivation with VLA-Rolling.
To provide a direct comparison, we adapted the official STEP implementation to π0.5 and evaluated both methods on LIBERO, with 10 episodes per task. For a fair comparison, we limited VLA-Rolling to a maximum denoising budget of 2 steps, matching STEP's denoising budget. The results are summarized below.
For training, STEP was adapted to π0.5 and trained for 100K steps. Under the same GPU training-time budget, we retrained VLA-Rolling for approximately 48K steps. Despite requiring fewer training steps, VLA-Rolling achieves a higher success rate and lower inference latency under the evaluated setting.