- Rating
- 2(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper presents ASSIST-VLA, a framework for accelerating VLA inference without additional training. It combines cognition caching based on action consistency, visual token selection using semantic and temporal cues, and structural MLP pruning guided by action influence. Experiments on CogACT in SimplerEnv report a 1.45× speedup with average success close to the original policy.
Critical strengths
-
The paper reports measured end-to-end acceleration. Latency decreases from 229.1 to 158.5 ms, including online module overhead, providing practical evidence beyond theoretical FLOPs reductions.
-
The ablations reveal the performance and efficiency trade-offs. Removing individual modules quantifies their marginal effects, and the appendix provides detailed cache rules, token selection procedures, and pruning configurations.
Critical weaknesses
-
The distinctive mechanisms are insufficiently isolated from existing acceleration techniques. EfficientVLA already combines multiple acceleration strategies. The proposed action influence score is not compared with simpler pruning criteria under the same budget, and action guided caching lacks a comparison with periodic reuse at a matched refresh rate.
-
The evaluation is narrow and does not establish robust performance preservation. Results cover one backbone and four task categories without reported uncertainty. Average success hides a 5.7 point decline on Pick Coke Can, while removing cognition reuse improves average success from 67.8% to 72.7%. Since reuse bypasses current visual observations, robustness to unexpected changes remains untested.
Questions for authors
-
What benefits distinguish the proposed framework from EfficientVLA beyond combining different implementations of caching, token selection, and model compression?
-
At the same pruning budget, how does the proposed action influence score compare with weight magnitude, activation statistics, and simpler action sensitivity criteria?
-
Does action consistency outperform periodic or random caching at the same refresh rate? How does it respond to unexpected object motion or failed grasps during reuse steps?
-
Could the authors provide latency versus success curves and repeated-run confidence intervals, and validate the method on another VLA backbone?
4 comments
Response to Your Questions (Part 2/2): Cognition Reuse and Evaluation Uncertainty
This is Part 2 of our response, addressing Q3–Q4. Part 1 discusses the differences from EfficientVLA and the pruning comparisons.
Q3. Does action consistency outperform periodic/random caching at matched refresh rates? How does it handle disturbances?
Action-consistency gating allocates reuse to segments where historical action plans remain stable. Periodic caching schedules reuse by timestep; random caching samples reuse steps. Our gate instead checks whole-plan, prefix, shifted-plan, first-action, and gripper consistency. Even at the same average refresh rate, these strategies can allocate computation to different steps.
Existing logs covering 300 PickCan, 216 Drawer, 108 DrawerApple, and 24 Carrot episodes show:
| Task | Cache hits | Control steps | Hit rate |
|---|---|---|---|
| PickCan | 5,630 | 24,000 | 23.46% |
| Drawer | 5,502 | 24,408 | 22.54% |
| DrawerApple | 4,710 | 21,600 | 21.81% |
| Carrot | 518 | 2,880 | 17.99% |
| Total | 16,360 | 72,888 | 22.45% |
Although at most one consecutive reuse step is permitted, actual hit rates are only 17.99%–23.46%; remaining steps perform full refreshes. Reuse frequency therefore depends on both consistency conditions and control trajectories.
After one reuse step, the next step must refresh, allowing new visual information to condition action generation. A failed consistency check exits the stable state; two consecutive qualifying evaluations are required to re-enter it. Reuse itself does not read new images, so unexpected object motion or failed grasps can first affect visual conditioning at the next refresh. This limits consecutive observation skipping to one step.
Across three seeds, complete ASSIST-VLA and the variant without reuse achieve episode-weighted success of 78.19% and 78.14%, and four-task averages of 69.88% and 68.86%. The original single-run decrease was not reproduced. Enabling reuse reduces measured latency from 188.7 to 158.5 ms.
Time constraints prevented matched-refresh-rate periodic/random comparisons and controlled disturbance experiments.
Q4. Confidence intervals, latency–success curves, and another backbone
We evaluated the baseline, complete ASSIST-VLA, and the variant without reuse across three seeds. Each seed includes 300/216/108/24 PickCan/Drawer/DrawerApple/Carrot episodes: 648 per seed and 1,944 per configuration.
Success rates (%) are:
| Method/seed | PickCan | Drawer | DrawerApple | Carrot | Overall |
|---|---|---|---|---|---|
| Baseline/1 | 92.33 | 73.61 | 42.59 | 45.83 | 76.08 |
| Baseline/2 | 95.00 | 68.06 | 44.44 | 41.67 | 75.62 |
| Baseline/3 | 92.33 | 73.15 | 54.63 | 33.33 | 77.47 |
| ASSIST/1 | 87.00 | 79.63 | 69.44 | 50.00 | 80.25 |
| ASSIST/2 | 85.00 | 77.31 | 53.70 | 54.17 | 76.08 |
| ASSIST/3 | 83.67 | 74.07 | 78.70 | 45.83 | 78.24 |
| No reuse/1 | 87.67 | 79.63 | 62.04 | 45.83 | 79.17 |
| No reuse/2 | 85.00 | 77.78 | 59.26 | 54.17 | 77.16 |
| No reuse/3 | 83.33 | 74.54 | 79.63 | 37.50 | 78.09 |
Across-seed means and 95% Student’s t-confidence intervals (%) are:
| Metric | Baseline | ASSIST | No reuse |
|---|---|---|---|
| PickCan | 93.22 [89.40,97.05] | 85.22 [81.05,89.39] | 85.33 [79.90,90.76] |
| Drawer | 71.60 [63.95,79.26] | 77.01 [70.07,83.94] | 77.31 [70.91,83.72] |
| DrawerApple | 47.22 [31.12,63.32] | 67.28 [35.89,98.68] | 66.98 [39.53,94.42] |
| Carrot | 40.28 [24.47,56.09] | 50.00 [39.65,60.35] | 45.83 [25.13,66.53] |
| Average | 63.08 [61.36,64.81] | 69.88 [64.73,75.03] | 68.86 [68.46,69.27] |
| Overall | 76.39 [73.99,78.78] | 78.19 [73.01,83.37] | 78.14 [75.64,80.63] |
Average weights tasks equally; Overall weights episodes. ASSIST exceeds baseline on DrawerApple in every seed, with pooled success of 67.28% versus 47.22%. PickCan remains lower at 85.22% versus 93.22%. These results show task-specific trade-offs and repeated-run uncertainty.
Latency–success curves require closed-loop evaluations and latency measurements across cache thresholds, token budgets, and pruning ratios. Time constraints prevented this systematic sweep.
Second-backbone validation requires adapting the reusable representation, action interface, and channel-influence mappings, followed by closed-loop evaluation. This validation has not yet been completed. We prioritized the direct pruning comparisons and repeated evaluations to examine action-influence scoring and cognition reuse.
Thank you for the additional experiments and clarifications. Your response addresses some of my concerns, but I still have three questions.
-
How did you get 57.97% on Drawer for “J-space only 1” in Q2? Did you use the same 216 episodes as in Q4? If so, how many were successful?
-
At the same pruning budget, J-space alone gets 64.62% average success, while the combined score gets 62.60%. What led you to choose the combined score?
-
If you reuse the cache at regular intervals, match the hit rate for each task, and avoid reuse on consecutive steps, how do success rate and latency compare with your action-consistency gate?
Thank you for examining the experimental statistics further and raising these three specific questions. We rechecked the episode counts, completed the missing Drawer evaluations, and reviewed our existing action-consistency gating experiments.
Question 1: How was the 57.97% Drawer result for J-space only calculated? Did it use the same 216 episodes as Q4?
Like Q4, this experiment was planned for 216 Drawer episodes. Nine episodes did not finish because of GPU memory errors during model loading. Of the 207 completed episodes, 120 succeeded, giving 120/207 = 57.97%. We apologize for not stating the number of completed episodes, which made the denominator ambiguous.
We have now rerun the nine missing episodes, obtaining three successes and six failures. The complete result is 123/216 = 56.94%, and the four-task average for J-space only changes from 64.62% to 64.37%.
| Pruning criterion | PickCan | Drawer | DrawerApple | Carrot | Four-task average |
|---|---|---|---|---|---|
| J-space only | 90.33% | 56.94% | 60.19% | 50.00% | 64.37% |
| Combined action influence and activation | 93.00% | 69.44% | 37.96% | 50.00% | 62.60% |
We will correct these values in the revised manuscript and report both successful and evaluated episode counts.
Question 2: Why choose the combined score when J-space only has a higher four-task average?
The combined score considers both a channel’s influence on the final action and its activation on actual inputs. Our additional closed-loop experiments show that these factors correspond to different task outcomes. Activation-based and weight-magnitude pruning largely preserve performance on PickCan and Drawer, but achieve only 25.93% and 18.52% on DrawerApple, respectively, below the unpruned baseline’s 41.67%. The simpler action-sensitivity criterion, which uses only our J-space score, achieves 60.19% on DrawerApple but drops to 56.94% on Drawer, compared with 73.61% for the baseline.
The combined score achieves 93.00%, 69.44%, 37.96%, and 50.00% on PickCan, Drawer, DrawerApple, and Carrot, respectively. Among these pruning comparisons, its per-task success-rate profile is closest to the unpruned baseline. This motivates our choice: to compress the MLPs while preserving the original model’s performance across tasks as closely as possible.
We also acknowledge that J-space only has a higher four-task average: 64.37% versus 62.60%. It performs better on DrawerApple and is a competitive, simpler alternative. The combined score performs better on PickCan and Drawer and more closely follows the baseline’s per-task results. We will present this trade-off in the revised manuscript and narrow our claim about the combined score’s advantage accordingly.
With only the combined-score MLP pruning enabled, DrawerApple success is 37.96%. Complete ASSIST-VLA reaches 69.44%, 53.70%, and 78.70% across three random seeds, exceeding the corresponding unpruned baseline in each seed. The closed-loop results with visual selection and MLP pruning together also show why the pruning criterion should be evaluated within the complete inference pipeline.
Question 3: How does action-consistency gating compare with fixed-interval reuse at a matched hit rate, with consecutive reuse prohibited?
This comparison asks which control steps should reuse cached information under a similar computation budget. We reviewed our existing exploratory DrawerApple experiments with cognition reuse only, using 20 episodes per configuration:
| Configuration | Success rate | Cache hit rate | Mean latency |
|---|---|---|---|
| Full refresh at every step | 20/20 (100%) | 0% | 219.9 ms |
| Maximum refresh interval of 3 | 0/20 (0%) | 38.5% | 158.7 ms |
| Maximum refresh interval of 2 | 20/20 (100%) | 22.0% | 184.1 ms |
| Maximum interval of 3; require three consecutive stability checks | 0/20 (0%) | 24.0% | 182.1 ms |
| Maximum interval of 3; raise the stability threshold to 0.92 | 0/20 (0%) | 25.0% | 178.7 ms |
The configurations with 22.0% and 24.0% hit rates have similar mean latencies (184.1 ms and 182.1 ms) but success rates of 100% and 0%, respectively. Lowering the average hit rate alone did not restore task performance; the timing of reuse and the length of consecutive observation skips also affect the result.
The 22.0% hit-rate configuration permits at most one reuse step before a forced refresh. It retained the baseline’s 20/20 successes while reducing mean latency from 219.9 ms to 184.1 ms, a reduction of approximately 16.3%. These results provide empirical motivation for action-consistency scheduling. Because the configurations also change gating parameters, they do not isolate the contribution of the consistency test itself. We will add the proposed direct comparison with fixed-interval reuse at a matched actual hit rate and with consecutive reuse prohibited, reporting per-task success, actual hit rate, and end-to-end latency.
Response to Your Questions (Part 1/2): Method Differences and Pruning Comparisons
We thank you for recognizing the measured acceleration, ablations, and implementation details. Due to the character limit, we divide our response into two parts. This comment addresses Q1–Q2; Part 2 addresses Q3–Q4.
Q1. What distinguishes ASSIST-VLA from EfficientVLA?
ASSIST-VLA introduces action consistency, semantic–spatiotemporal similarity, and J-space action influence to determine when computation is needed, which visual information to retain, and which model channels to preserve.
Action consistency connects reuse decisions to control outputs. During continuous manipulation, observations may change while required actions remain stable. We check whole-plan consistency, short-horizon prefixes, shifted-plan alignment, and gripper commands to decide whether cognition needs recomputation. Each hit avoids visual encoding, projection, and the entire multimodal language backbone, while retaining action-head execution. This exploits continuity across control steps rather than repetition within diffusion generation.
Semantic–spatiotemporal similarity preserves static targets and dynamic interaction regions. Motion-only selection can miss stationary targets, while single-frame importance cannot exploit observation changes. We combine instruction relevance with local temporal correlations and neighborhood displacement responses, allowing the visual budget to cover targets, the end effector, and interaction regions. This shortens the language backbone’s input on refresh steps.
J-space evaluates channels through their influence on actions. We compose layer-to-cognition and cognition-to-action mappings and combine the resulting influence with activation information. Ranking therefore considers both channel usage and its effect on control outputs. Physically reconstructing narrower gate/up/down projections converts this analysis into dense-computation and storage savings.
Together, these mechanisms reduce both refresh frequency and per-refresh cost. In the reported CogACT adaptation comparison, ASSIST-VLA achieves 158.5 ms latency and 67.8% success, versus EfficientVLA’s 168.5 ms and 66.3%: 5.9% lower latency and 1.5 points higher success. Relative to CogACT’s 229.1 ms, our speedup is 1.45×.
Removing cognition reuse, visual selection, or MLP pruning increases latency to 188.7, 171.1, or 178.4 ms, respectively. These ablations connect the mechanisms to measured end-to-end savings.
Q2. How do pruning criteria compare at the same budget?
We added closed-loop evaluations with MLP pruning only at the same overall pruning ratio. Success rates (%) and unweighted four-task averages are:
Activation and magnitude pruning largely preserve PickCan and Drawer performance, but reduce DrawerApple success to 25.93% and 18.52%, versus the unpruned baseline’s 41.67%. Their scores alone do not fully capture channel contributions across tasks.
The simple action-sensitivity controls use only our J-space action-influence score, without activation information. They achieve 60.19% and 58.33% on DrawerApple under two settings, both above baseline. However, the first configuration reduces PickCan and Drawer to 90.33% and 57.97%. J-space thus identifies useful action-related channels, while using it alone changes the task profile.
Our complete criterion combines action influence and activation information. Among controls evaluated on all four tasks, its per-task success profile is closest to the unpruned baseline. Its average exceeds activation and magnitude pruning by 1.91 and 3.41 points, respectively. These closed-loop measurements evaluate task performance independently of the channel-selection score.