- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper introduces MARS, a training-free, inference-time safety intervention framework designed for frozen Vision-Language-Action (VLA) models based on flow matching. The method first employs "Grounded Risk Verbalization" (GRV) to identify scene-specific risks using a vision-language model, translating undesirable interactions into counterfactual action instructions. Subsequently, through "Trajectory-Anchored Flow Repulsion" (TAFR), the framework compares the velocity fields—derived from the original task instructions versus the counterfactual instructions—along a shared nominal sampling trajectory. The resulting clipped and weighted residuals are aggregated into a corrective term, which is then used to adjust the final action sequence. The authors evaluate MARS across three simulated safety benchmarks, four VLA backbones, and two real-world robotic platforms, while also providing analyses on component ablation, hyperparameter sensitivity, and computational overhead. The results indicate that, when applied to the π0.5 model, MARS simultaneously improves both task completion rates and safety; however, its benefits are more limited when applied to other backbones, in some cases even leading to performance degradation.
Critical strengths
This paper focuses on protecting surrounding objects and hazardous zones during task execution, explicitly distinguishing between task success and safe execution—a distinction that aligns with the practical requirements of deploying general-purpose robots.
MARS does not require updating VLM or VLA parameters, nor does it rely on explicit geometric models or separately trained evaluators. Its overall design is clear, bridging semantic risk identification with continuous action correction through counterfactual action instructions.
By decoupling nominal sampled trajectories from cumulative corrections, the approach prevents the correction terms from continuously shifting the evaluation points within the subsequent velocity field. The authors provide experimental evidence to support this design by comparing it against an approach that integrates corrections into the state.
The paper evaluates the method across three simulation benchmarks, four model backbones, and two real-world robotic platforms, while analyzing factors such as prompting strategies, clipping magnitudes, guidance strength, and the frequency of VLM invocations. The authors also discuss the limitations of the method when applied to less capable model backbones.
Critical weaknesses
Appendix J suggests that modifying the state integral invalidates the upper bound on the cumulative correction derived in Equation (14); however, a similar upper bound can still be derived for the cumulative correction itself, provided that dimension-wise clipping and weight constraints are maintained. It is important to distinguish between the cumulative correction and the total deviation of the final output from the original nominal trajectory. Furthermore, the fact that deviations in the action space are bounded does not, in itself, guarantee physical safety.
The collision rate reported in Table 7 is 162.50%, which exceeds the range permitted by its own definition. Additionally, the description of the numerical values and the direction of change for the prompt-injection method "CC" (in Appendix H.1) is inconsistent with the data in Table 7. These issues undermine the credibility of the conclusions drawn from the relevant experiments.
The primary comparisons focus on the original policy, prompt injection, and internal integral variants. The absence of direct comparisons with other safety-filtering methods or simpler action-space correction schemes makes it difficult to fully substantiate the distinct advantages of TAFR.
The prompt for GRV requires the exclusion of "operation targets"; however, in "careful grasping" tasks, the hazardous regions are often part of the target object itself. The paper does not clearly explain how the current object-level risk representation identifies and avoids component-level risks—such as the metal shaft of a screwdriver.
The latency results presented in Tables 10 and 11 are difficult to reconcile with the described parallel execution mechanism. Moreover, the primary safety gains are concentrated in the setting, while other backbone models fail to show consistent improvement; thus, the current results are insufficient to support a claim of broad, generalizable effectiveness.
The main results lack confidence intervals or inter-seed variance. Given that each experiment consists of only 20 trials per task for each method, it is difficult to determine whether the observed performance differences are stable.
Questions for authors
- Why does Equation (14) no longer hold for the cumulative correction , given that the same clipping and weight constraints are applied to the correction-state integral? Is the authors' intent to constrain this cumulative correction, or rather the total output deviation relative to the unguided policy?
- Please verify the 162.50% CR value in Table 7, and address the discrepancy between the CC values in the table and those in Appendix H.1. Following any necessary corrections, does the explanation regarding performance degradation due to prompt injection still hold?
- Could the authors include a baseline—such as a simple action-correction method—that utilizes the same risk information and a comparable computational budget, along with an appropriate geometric safety-filtering approach, to clearly demonstrate the incremental benefits of TAFR?
- In the "cautious grasping" task, how does GRV represent the hazardous regions of a target object itself? Please provide examples of actual risk records and counterfactual instructions, and clarify whether task-specific rules are employed.
- Are the timing conditions consistent between Table 10 and Table 11? Please explain the discrepancy (102 ms vs. 114 ms) when , as well as why the latency reaches 193 ms when under parallel execution.
- Could the authors provide confidence intervals based on multiple random seeds, paired statistical analyses for real-world experiments, and an assessment of the impact of risk-cache lag in dynamic scenarios? Furthermore, for backbones that exhibit unstable performance, are there established criteria for enabling or rolling back these interventions?
10 comments
1. Clarification on the Theoretical Bound in Appendix I & J (W1 & Q1)
Q1: Why does Equation (14) no longer hold for the cumulative correction , given that the same clipping and weight constraints are applied to the correction-state integral? Is the authors' intent to constrain this cumulative correction, or rather the total output deviation relative to the unguided policy?
Thank you for pointing out this distinction. You are right on both points, and the submitted Appendix J was imprecise. It stated that the bound of Equation (14) "no longer holds" for the cumulative correction in the corrected-state variant. That statement is wrong as written: with the same component-wise clipping and weights applied to , the telescoping argument of Appendix I applies verbatim to , so holds in both variants. What fails in the corrected-state variant is the identity , not the bound on .
-
What Equation (14) is meant to bound. Our intent is to bound the total output deviation of the final action chunk relative to the unguided nominal policy, , where is the endpoint of the unguided ODE integrated from the same initial noise.
- TAFR evaluates all velocities on the unperturbed anchor trajectory, so and independent of the Lipschitz properties of .
- In the Corrected-State Variant (Appendix J), evaluating at the shifted state alters every subsequent nominal velocity evaluation, so the terminal state itself drifts away from and The second term is still at most . The first term is not controlled by Equation (14). Writing and letting denote the -Lipschitz constant of in its state argument (uniform over ), the Euler recursion gives , and since , The corrected-state variant is therefore not unbounded, but its guarantee is rather than . Because scales as near the end of integration, is large and not known a priori, which makes this bound vacuous in practice.
- This is also what the trajectory-anchoring ablation measures empirically: on VLA-Arena with , the corrected-state variant reaches 43.8% SR versus 49.2% for TAFR (Figure 5(b), Section 4.3; Table 13, Appendix H.7), consistent with the drift of the first term.
-
Action-Space Bound vs. Physical Safety. We agree that bounding action-space deviation does not, by itself, guarantee physical safety. Equation (14) is an action-locality guarantee: it states that the steered chunk lies within a hyperparameter-controlled neighborhood of the nominal chunk in the backbone's normalized action coordinates. Whether that neighborhood is collision-free depends on the backbone's action scale and on the scene. We have revised Section 3.2, Appendix I, and Appendix J to (i) correct the misstatement about , (ii) add the comparison as a new equation in Appendix J, and (iii) describe Equation (14) as action locality rather than safety.
2. Correction of the Transcription Typo in Table 7 (W2 & Q2)
Q2: Please verify the 162.50% CR value in Table 7, and address the discrepancy between the CC values in the table and those in Appendix H.1. Following any necessary corrections, does the explanation regarding performance degradation due to prompt injection still hold?
We sincerely apologize for this proofreading oversight in Table 7 and thank you for catching it. You are right that 162.50% is invalid for a collision rate in and that 35.82 in Table 7 directly contradicted our text six lines below in Appendix H.1.
-
Verification & Internal Manuscript Cross-Checks: Across all 5 evaluation seeds, the true values for Prompt injection are CC = 5.82 ± 0.58 on VLA-Arena and CR = 62.50 ± 1.76% on SafeLIBERO:
- Cause of the typo: During table transcription, stray leading digits were inadvertently typed into the two safety-cost cells (5.82 → 35.82 and 62.50 → 162.50), and a subsequent column-wise formatting pass (bold for best, underline for second-best) mechanically underlined 12.55 and 79.38 based on those mistyped numbers.
- Consistency with the submitted manuscript: Notably, the submitted prose in Appendix H.1 (six lines below Table 7) already reported the true value—"The lower CC (12.55→5.82) confirms that the robot largely fails to initiate meaningful motion rather than completing tasks more safely"—and Figure 5(b) in Section 4.3 plotted the exact same Prompt injection runs (15.7% VLA-Arena SR, 22.6% LIBERO-Safety SSR, and 7.5% SafeLIBERO SSR).
-
Corrected Table 7 with 5-Seed Statistics: Below we report the corrected Table 7 together with SafeLIBERO SR:
| Strategy ( Backbone, 5 Seeds) | VLA-Arena SR () | VLA-Arena CC () | LIBERO-Safety SSR () | SafeLIBERO SR () | SafeLIBERO SSR () | SafeLIBERO CR () |
|---|---|---|---|---|---|---|
| No VLM guidance | 39.3 ± 2.3% | 12.55 ± 0.91 | 56.4 ± 2.5% | 59.69 ± 1.74% | 18.75 ± 1.35% | 79.38 ± 1.47% |
| Prompt injection | 15.7 ± 1.5% | 5.82 ± 0.58 | 22.6 ± 1.9% | 24.38 ± 1.52% | 7.50 ± 0.93% | 62.50 ± 1.76% |
| Counterfactual (GRV, Ours) | 49.2 ± 2.4% | 7.53 ± 0.68 | 60.8 ± 2.1% | 61.44 ± 1.78% | 21.25 ± 1.51% | 47.63 ± 1.93% |
- Does the Explanation in Appendix H.1 Still Hold?
Yes—correcting the table to CC = 5.82 and CR = 62.50% directly restores the explanation in Appendix H.1 (which had been drafted from the true 5.82 value):
- Appending explicit negative avoidance clauses (e.g., "pick up the bowl and place it on the plate, while avoiding the wine bottle") produces compound instructions with linguistic negation that lie outside 's single-objective fine-tuning distribution.
- Just as SmolVLA in Tables 1 and 4 incurs low CC (8.43) only because of very low task completion (18.0% SR), Prompt injection's lower raw contact count (CC = 5.82, CR = 62.50%) reflects motion failure rather than safe execution (SR drops from 39.3% → 15.7% on VLA-Arena and 59.69% → 24.38% on SafeLIBERO; SSR drops from 18.75% → 7.50%). By contrast, Counterfactual (GRV) + TAFR keeps each prompt as an in-distribution, single-objective affirmative command (e.g., "pick up the wine bottle") and applies repulsion in velocity space, simultaneously achieving the highest task completion and safe success rate (49.2% VLA-Arena SR, 60.8% LIBERO-Safety SSR, 61.44% SafeLIBERO SR / 21.25% SSR) and the lowest overall collision rate (47.63% SafeLIBERO CR). We have fixed Table 7 in Section 4.3 and Appendix H.1.
We thank Reviewer Ei1G for the thorough and constructive review. Your questions on the theoretical bound, the Table 7 typo, the missing baselines, and the CautiousGrasp scope were all well-placed, and the paper is improved as a result. We address each point below.