- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes FlatVLA, which combines Sharpness-Aware Minimization (SAM) with sequential weighting of action prediction losses to improve VLA robustness. The authors theoretically connect parameter-space flatness to an upper bound on compounding error. Experiments on four VLA models, LIBERO, LIBERO-Plus, and real-world robotic tasks demonstrate improvements in performance and robustness.
Critical strengths
The paper attempts to explain closed-loop robustness in VLAs from the perspective of parameter-space flatness while considering the sequential structure of action prediction. This is an interesting and potentially significant research direction. The method requires no changes to the original model architecture or inference procedure.
Critical weaknesses
-
Prior work cited in the paper has already applied SAM to VLA training [1]. The main algorithmic addition in FlatVLA is Sequential Weighting, and the two methods share much of their core optimization approach. The paper provides neither a direct experimental comparison with [1] nor ablations of standard SAM and Sequential Weighting alone. The independent contribution of the proposed weighting strategy remains unclear.
-
Proposition 1 relies on additional sensitivity assumptions, while Theorem 1 only establishes a comparison of upper bounds on compounding error. The connection between episode-level error propagation and action-chunk-level weighting, as well as the choice of a fixed , is not sufficiently justified.
-
Although the success rates improve, the loss landscape, predictive uncertainty, and action smoothness analyses do not directly demonstrate reduced policy sensitivity or compounding error. The suggested connection between increased uncertainty in failed executions and recovery potential also lacks direct evidence.
Questions for authors
-
What is the main methodological contribution of FlatVLA compared with the cited work [1]? Why is there no direct experimental comparison? Could the authors provide ablations of standard SAM, Sequential Weighting alone, and the full FlatVLA method to clarify the benefits of the proposed weighting strategy?
-
Could the authors further explain the applicability of Assumptions 4 and 5, and how Eq. (4) supports the weighting design in Eq. (9)? Is there experimental evidence supporting the choice of ?
-
Could the authors directly measure policy sensitivity and error propagation under perturbations to validate the theoretical analysis and the claimed robustness improvement?
[1] Flatness Preserves Instruction Following in Vision-Language-Action Models.
5 comments
Response to Reviewer DH2q (Part 2/2)
W2, Q2 Part 1: Theoretical Framework and Sequential Weighting
Our analysis links local loss sharpness, policy sensitivity, and sequential error propagation. Assumptions 4–5 specify the local conditions under which Proposition 1 relates sharpness to policy sensitivity, while Theorem 1 conditionally compares compounding-error upper bounds without guaranteeing lower realized errors. Sequential weighting in Eq.(9) is motivated by, rather than directly derived from, temporal error propagation in Eq.(4): during open-loop execution, earlier action errors can affect subsequent physical states while later actions are not recomputed.
We acknowledge that is a fixed design hyperparameter, not an estimated dynamics constant. We set for OpenVLA-OFT () and (), assigning 61.6% and 60.4% of the normalized weight to the first half of each chunk, respectively. For the longer chunks of (), we use a more conservative , assigning 51.0% to the first half. These settings adjust the emphasis on earlier actions according to the predicted action horizon.
To test the novelty of FlatVLA, we evaluated replanning intervals { } with , which has the longest predicted action horizon. Here, FlatVLA consistently shows improvement across all approach, providing empirical support of our apporach.
| Replan () | Baseline | SAM only | SW only | FlatVLA |
|---|---|---|---|---|
| 93.3 | 93.5 | 93.5 | 94.3 | |
| 88.1 | 89.1 | 89.2 | 91.0 | |
| 83.4 | 84.3 | 83.6 | 87.3 | |
| 68.7 | 70.1 | 69.9 | 75.6 |
We provided these explnataions and complete results for above experiments in the revised manuscript.
W3, Q2 Part 2, Q3: Policy Senstivity and Error Propagation
-
Policy Sensitivity: We distinguish training-loss sharpness from execution-level policy robustness. Since VLA models are trained with action-supervised objectives, training-loss sharpness provides an optimization-level perspective on the learned policy. To quantify this, we measure local loss sharpness as . As shown below, FlatVLA achieves the lowest measured sharpness within each backbone.
Backbone Baseline SAM only SW only FlatVLA Reduction vs. Baseline OpenVLA 0.2143 — — 0.0707 67.0% OpenVLA-OFT 0.0102 0.0081 0.0131 0.0038 62.7% 0.1631 0.0552 0.1541 0.0187 88.5% 0.1325 0.0322 0.1539 0.0108 91.8% Both Fig. 5 and our quantitative sharpness results point to reduced loss sensitivity (Assumption 4), whereas the robustness gains under multimodal perturbations on LIBERO-PLUS substantiate policy sensitivity (Assumption 5). This empirical evidence further validates our theoretical connection between loss sharpness and policy sensitivity.
-
Error Propagation: To examine the effect of early action errors, we conduct additional experiments by injecting action noise during the first 20 rollout steps (1–20) or the next 20 steps (21–40) on . As shown below, early-stage perturbations cause substantially greater performance degradation than later-stage perturbations, consistent with the greater downstream influence of earlier action errors. As results, FlatVLA exhibits the smallest performance drops under both settings, providing further evidence of its robustness to temporally localized action perturbations.
Noise Injection () Baseline SAM only SW only FlatVLA No Noise 92.2 92.9 92.5 94.3 First 20 steps (1–20) 59.9 (−32.3 pp) 63.3 (−29.6 pp) 60.5 (−32.0 pp) 67.5 (−26.8 pp) Next 20 steps (21–40) 87.4 (−4.8 pp) 88.8 (−4.1 pp) 87.1 (-5.4pp) 90.7 (−3.6 pp) Here, () indicates performance drop pp relative to No Noise.
In last, we revised the interpretation of predictive uncertainty (Table 5), clarifying that the observed success–failure uncertainty gap does not directly demonstrate recovery capability.
Brief Reminder for Reviewer DH2q
Dear Reviewer DH2q,
Thank you again for your time and feedback. We hope our responses and the updated manuscript (e.g., the additional experiments and theoretical clarifications) have adequately addressed the questions and concerns raised. For the convenience,, we have also posted an overall summary of where each revision was made in our general response at the top. As the end of the discussion period approaches, we would be glad to respond if any further questions or concerns remain.
Best regard,
The Authors
Response to Reviewer DH2q (Part 1/2)
We thank the reviewer for the constructive feedback and address the novelty concerns below. We have incorporated additional analyses and theoretical clarifications into the revised manuscript. For convenience, all changes are highlighted in orange.
W1, Q1: Contribution beyond SAM and Ablations
FlatVLA differs from prior work [1] in both its contribution and robotics-specific motivation. Unlike prior work [1], FlatVLA goes beyond applying SAM by incorporating the sequential structure of action chunks to account for error propagation during execution. We also provide an intuitive theoretical analysis linking loss sharpness, policy sensitivity, and sequential error propagation under stated conditions.
We additionally compare the baseline as follows: (1) the original baseline, (2) SAM only (γ=0; partial reproduce of [1]), (3) sequential weighting without SAM (SW only), and (4) FlatVLA The result shows that FlatVLA consistently improves in all cases, with gains across all backbone and methods.
LIBERO
LIBERO-PLUS
We provided full LIBERO and LIBERO-PLUS results and the corresponding revisions are included in the revised manuscript.
[1] Zhang and Bisk, Flatness Preserves Instruction Following in Vision-Language-Action Models Arxiv'26