- Rating
- 3(+1)
- Confidence
- 3(+1)
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3(+1)
Read review
Paper summary
This paper studies how a specified visual perturbation changes the executed-action distribution of a flow-matching VLA at a fixed observation. It propagates the measured context change through clean-trajectory Jacobians and bounds the nonlinear remainder using recursive curvature tubes. A Wasserstein bound is then converted into a clean-action retention bound, with a refinement accounting for predicted mean translation. Experiments with a LIBERO-finetuned π0.5 policy examine six perturbation families at two observations. They show directional differences at matched context magnitudes, close numerical agreement for small perturbations, and increasing conservatism at larger magnitudes.
Critical strengths
The recursive tube construction provides a coherent way to control unknown perturbed trajectories using the clean trajectory and measured context change. I could follow the containment argument in Equations (14)–(16) and Appendix A.1.1. The analysis also explicitly distinguishes generated chunks from executed coordinates. Its assumptions and the distinction between exact bounds and numerical estimates are clearly acknowledged.
The experiments support the importance of perturbation direction. At context magnitude 500, patch noise produces a sixfold larger empirical shift than contrast for the black-bowl observation, while their ordering reverses for cream cheese (Table 1). Translation adjustment also improves the retention estimates: for contrast near magnitude 180, the estimate increases from 22.7% to 68.7%, against observed coverage of 95.2% (Table 3). Separate fitting, calibration, and evaluation samples are used for the clean-action region.
Critical weaknesses
-
The reliability of the numerical remainder control is not established. The theoretical construction requires curvature upper bounds and enclosing tubes. Appendix A.2.1 acknowledges that finite searches and power iteration can underestimate the required quantities, so the computed tubes may fail to contain the actual displacements. The numerical study does not check this containment or report sensitivity to the search budgets. Agreement with empirical Wasserstein distance does not establish control of the shared-noise prediction error, since that distance can be smaller than the displacement being bounded. Moreover, the candidate-set construction and optimization procedure are not specified on pages 15–16. The timing uses five interpolation points, whereas the main-study description specifies 16. These omissions prevent me from assessing whether the reported tight estimates survive more thorough numerical evaluation.
-
The experiments do not isolate the value of recursive curvature control beyond first-order sensitivity. Figure 3 does not separately report the first-order term in Equation (17), the estimated remainder, and the actual linearization error. The unrestricted joint-direction comparison also changes the evaluation domain and pools maxima across 16 noise draws, so it does not isolate the effect of discarding direction. This matters because the estimates closely match empirical distances for very small perturbations but can become extremely loose: cream-cheese patch warp gives 9.5 against an empirical distance of 0.070 at magnitude 250 (Table 2). The retention application is encouraging, but its KDE bandwidth and explicit margin-lower-bound calculations are also missing. Thus I cannot verify the probability estimates or determine where the additional curvature calculation provides useful control.
Questions for authors
-
On the existing paired trajectories, does the actual prediction error remain within the numerical radius at every flow step? How does containment change when increasing displacement candidates, interpolation points, and power iterations? Please specify the search algorithms and configurations used for each plot and the timing.
-
Could you show the first-order term, estimated nonlinear contribution, and actual linearization error together for the existing observations? Please also clarify how differences in evaluation domain and aggregation across noise draws affect Figure 3.
-
What KDE bandwidth and explicit formulas, constants, and grid parameters produce the margin lower bounds in Appendix A.3? These are needed to reconstruct the retention estimates, including the 82% value in Figure 4.
4 comments
Response to t1KM: Q2
[Q2] Could you show the first-order term, estimated nonlinear contribution, and actual linearization error together for the existing observations? Please also clarify how differences in evaluation domain and aggregation across noise draws affect Figure 3.
Equation (17) provides a rigorous upper bound under the assumptions of Theorem 1 by controlling the nonlinear remainder. We agree that the first-order approximation is often numerically accurate in practice in many of the tested conditions in the low perturbation regime, although it can underestimate the measured displacement. Tables 6 and 7 in Appendix A.2.4 compare the first-order prediction, full estimate, and measured vector remainder using matched samples at magnitudes 25, 100, and 250. The magnitude-250 results are shown below.
Let and be the actual and first-order predicted executed-action displacements for the same initial-noise draw. We report the paired RMS displacement , its first-order prediction , and the vector error . Here is a matched-sample diagnostic, distinct from the empirical Wasserstein distance . Here is the full numerical estimate. Its nonlinear part is , which can be obtained directly from the table. Vector error includes both magnitude and direction differences, which a comparison of magnitudes alone can miss.
All quantities use the original eight curvature draws and ten power iterations. We retain the original nonlinear term and average the first-order term over these same eight draws. Table 9 uses 128 draws for its first-order average, which explains the small remaining differences in . The curvature search uses six directions and seven interpolation points for black bowl, and four directions and five interpolation points for cream cheese. The first-order prediction uses the measured context change and derivatives along the clean flow trajectory. All values are in normalized action units.
Black bowl,
| Perturbation | Paired RMS | First order | Full estimate | Vector error |
|---|---|---|---|---|
| Brightness | 0.12673 | 0.12116 | 0.21169 | 0.01157 |
| Contrast | 0.13339 | 0.13646 | 0.19846 | 0.00882 |
| Patch noise | 0.24917 | 0.24272 | 0.26712 | 0.01182 |
| Patch blur | 0.28707 | 0.28137 | 0.35533 | 0.02276 |
| Patch saturation | 0.18405 | 0.17369 | 0.41947 | 0.04357 |
| Patch warp | 0.19374 | 0.18699 | 0.26126 | 0.01526 |
Cream cheese,
| Perturbation | Paired RMS | First order | Full estimate | Vector error |
|---|---|---|---|---|
| Brightness | 0.05062 | 0.05148 | 0.06282 | 0.00233 |
| Contrast | 0.28601 | 0.28408 | 0.87702 | 0.02134 |
| Patch noise | 0.07539 | 0.07291 | 0.15560 | 0.00539 |
| Patch blur | 0.09032 | 0.09105 | 0.14577 | 0.00401 |
| Patch saturation | 0.10858 | 0.10984 | 0.16926 | 0.00418 |
| Patch warp | 0.07008 | 0.06905 | 9.51113 | 0.00470 |
At , first order underestimates by more than in 8 of 12 conditions. Across all 36 conditions in Tables 6 and 7, the full estimate covers and its nonlinear part covers . For black-bowl patch blur at magnitude 250, and , showing one example of informative nonlinear control even when the first-order magnitude is close.
Regarding Figure 3, now Figure 4 in the updated paper, our tube construction evaluates curvature along candidate directions, whereas the unrestricted joint-direction calculation searches arbitrary directions at clean trajectory points. These differences affect the resulting estimates. The convergence results in Appendix A.2.2 support the numerical stability of our tube estimates under substantially larger search budgets, complementing the figure’s evidence for the usefulness of retaining perturbation-direction information.
Response to t1KM: Q3
[Q3] What KDE bandwidth and explicit formulas, constants, and grid parameters produce the margin lower bounds in Appendix A.3? These are needed to reconstruct the retention estimates, including the 82% value in Figure 4.
The KDE underlying Figure 5 (Figure 4 in the original submission) can be reproduced using scipy.stats.gaussian_kde(X) with default settings, where X contains the first five principal-components of 2048 clean action samples. We use 2048 separate clean flow samples to estimate the 95%-coverage density threshold. We define the retained region by thresholding the negative log KDE density at a level calibrated for 95% clean coverage. For Figure 5, we compute margin lower bounds analytically using the density gradient and a curvature bound determined by the Gaussian kernel covariance. The displayed contours are two-dimensional projections of this five-dimensional region, and the retention estimates apply the margin and retention calculations in Appendix A.4 of the updated paper to this KDE. We have added pseudocode for the five-dimensional margin and retention calculation to Appendix A.4, including the threshold, curvature constant, and residual budget that gave the 82% reported in the main paper. We plan to release the code after paper decisions as well.
Rebuttal follow-up
Dear reviewer t1KM,
Thank you again for your helpful suggestions on numerical reliability and reproducibility. As the discussion deadline approaches, please let us know if our response has addressed your questions and concerns. We'd also be happy to provide further clarifications if you have any more questions. Thank you for your time.
Best,
The Authors
Response to t1KM: Q1
[Q1] Tube containment. How does containment change when increasing displacement candidates, interpolation points, and power iterations?
Thank you for raising this concern. We have also added the containment table and its evaluation protocol to Appendix A.2.3 of the revised paper, and reproduce the table below for convenience. Our tube construction bounds the nonlinear error around the first-order displacement prediction. Here we show that tube containment failures are infrequent and small in absolute magnitude given the numerical approximation methods used. We'd like to emphasize that theory guarantees tube containment. The true displacement can lie outside of the tube in practice due to numerical underestimation from finite searches and power iteration. The inequality below describes how we checked numerical tube containment with an absolute tolerance τ=10−5 in normalized full-state units:
∥δk−δ^k∥F≤r^k+τ,
where δk is the actual clean-to-perturbed displacement, δ^k its first-order prediction, and r^k the numerical radius.
For the two observations and six perturbation families in the main study, the paper settings use eight Gaussian initial-noise samples per observation, shared between clean and perturbed runs, and 10 power iterations. The table reports the percentage of checks that fail the above criterion. Each entry represents 96 checks: two observations × six perturbations × eight initial-noise samples.
Containment failures (%) with τ=10−5. Column headings 1–10 denote flow steps.
Across the nine tested magnitudes from ∥Δc∥F=5 to 500, 6 of 8,640 checks (0.069%) fail this criterion. These six violations are small in absolute magnitude and occur only at ∥Δc∥F=10 and 25 with maximum excess of 1.1×10−5 and 3.77×10−5 respectively.
Numerical convergence. We have added a numerical convergence study in Appendix A.2.2, with results and computation times measured on an NVIDIA RTX 5090 GPU in Tables 2–4. The theory guarantees containment under its assumptions. These experiments assess the accuracy and cost of its numerical implementation.
Table 2. At the paper's budget of 10 power iterations, median operator-norm errors are 1.37% for black bowl and 3.27% for cream cheese, with norm-estimation time about 33× lower than explicit Jacobian construction and SVD. At 100 iterations, the median errors decrease to 0.03% and below 0.01%.
Table 3. Increasing the direction count from six for black bowl and four for cream cheese to 66 changes sampled curvature by at most 0.037% and the final estimate B by less than 0.0013%.
Table 4. Increasing the interpolation grids from seven and five points, respectively, to 49 changes B by at most 0.0086% across the tested conditions.
Tables 3 and 4 cover both observations and all six perturbations at ∥Δc∥F=100, operator norms computed with 100 power iterations. The small changes under substantially larger searches support the paper's direction and interpolation budgets. Table 2 separately quantifies the power-iteration accuracy–cost trade-off, and direct containment results are reported above.
Clarifications on configurations used for the plots and timing. Thank you for pointing out the inconsistency between the interpolation counts in Appendix A.2.1 and the timing description. The stated value J=16 in the appendix was a mistake. The main perturbation study uses J=7 for black bowl and J=5 for cream cheese, now documented in Appendix A.2.2. These runs use eight initial-noise draws and ten power iterations. The curvature ablation uses J=5. We have corrected and clarified this mistake in Appendix A.2.1.