SPAR: Structural Pruning and Action Recovery for Vision-Language-Action Models

Shenli Zheng; Zhou Fang; Yuan Du; Yijiang Liu; Li Du

cs.ROOct 19, 2026

SPAR: Structural Pruning and Action Recovery for Vision-Language-Action Models

Authors: Shenli Zheng, Zhou Fang, Yuan Du, Yijiang Liu, Li Du

Organizations: Nanjing University, Nanjing, China · Nanjing University of Information Science and Technology, Nanjing, China

Abstract

The limited onboard computing resources of robots make the inference demands of vision-language-action (VLA) models costly to accommodate. Reducing these demands through aggressive compression, however, can degrade manipulation performance beyond what supervised fine-tuning alone recovers. We present SPAR (Structural Pruning and Action Recovery), which simultaneously prunes the hidden and feed-forward intermediate dimensions of the large language model (LLM) backbone through pruning-aware tuning, then applies reinforcement learning (RL) to recover policy performance. Applied to $π_{0.5}$ , SPAR reduces the language backbone from 2.34B to 0.09B parameters and the full model from 3.37B to 1.12B, removing approximately 96% and 67% of their parameters, respectively. At this compression level, RL raises average success across the four LIBERO suites from 84.55% to 93.45%, an 8.90-percentage-point gain. The compressed policy achieves a $1.98\times$ full-policy inference speedup over the unpruned model. These results identify substantial intermediate-dimension redundancy and show that post-pruning RL can recover manipulation performance in compact policies.

Figures & tables

Fig. 2: Overview of the SPAR Framework. Top: SPAR compresses the LLM backbone of a VLA policy, which integrates visual and language inputs to guide the action expert. Bottom: (1) Through pruning-aware tuning , SPAR jointly learns channel masks for the hidden dimension and the FFN intermediate dimension while adapting the policy to manipulation tasks. (2) These masks guide structural pruning , which removes inactive channels to form a smaller dense backbone. (3) After supervised fine-tuning of the compact policy, RL recovery further improves task performance through simulated interaction, updating the action expert and value head while keeping the pruned LLM backbone frozen.

Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Params (B) Total / LLM	Speedup ( $\uparrow$ )	Pruning Ratio		Spatial	Object	Goal	Long	Avg.
Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Params (B) Total / LLM	Speedup ( $\uparrow$ )	Model	LLM	Spatial	Object	Goal	Long	Avg.
$\pi_{0.5}$	baseline	100% / 100%	3.37 / 2.34	1.00	0%	0%	98.80	98.80	97.40	92.20	96.80
$\pi_{0.5}$	PAT only	100% / 90%	3.20 / 2.17	1.05	5%	7%	99.80	98.20	96.60	94.60	97.30
		100% / 70%	2.86 / 1.83	1.17	15%	22%	98.40	98.00	98.00	96.20	97.65
		100% / 50%	2.52 / 1.49	1.33	25%	36%	98.40	98.60	98.20	94.40	97.40
		100% / 30%	2.19 / 1.16	1.52	35%	51%	98.80	98.20	97.80	94.60	97.35

TABLE I: Effect of FFN Intermediate-Dimension Pruning on LIBERO Performance. We vary the FFN intermediate-dimension retention target in

\pi_{0.5}

while keeping the hidden dimension unchanged, and evaluate each policy after SPAR’s PAT stage without RL. The ratios

\rho_{hid}/\rho_{int}

specify retention targets for the hidden and intermediate dimensions. Parameter counts are in billions, and pruning ratios indicate the approximate percentages of parameters removed from the full model and LLM relative to the unpruned baseline. Success rates are reported as percentages, with Avg. denoting the mean across the four suites and bold highlighting the best score in each column.

Compact VLA Models
Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Params (B) Total / LLM	Speedup ( $\uparrow$ )	Pruning Ratio		Spatial	Object	Goal	Long	Avg. ( $\boldsymbol{\Delta}$ )
Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Params (B) Total / LLM	Speedup ( $\uparrow$ )	Model	LLM	Spatial	Object	Goal	Long	Avg. ( $\boldsymbol{\Delta}$ )
NORA [ 5 ]		–	3.00 / –	–	–	–	92.20	95.40	89.40	74.60	87.90
NORA-1.5 [ 6 ]		–	3.00 / –	–	–	–	98.00	96.00	95.40	90.50	95.00
SmolVLA [ 4 ]		–	2.25 / –	–	–	–	93.00	94.00	91.00	77.00	88.75
GraspVLA [ 7 ]		–	1.80 / –	–	–	–	–	94.10	91.20	82.00	89.10

TABLE II: LIBERO Performance of SPAR and Comparison Methods. PAT only denotes pruning and supervised adaptation, and PAT+RL adds RL recovery. Notation follows Table I . Parentheses show RL gains in percentage points. Bold marks the best success rates within each PAT group. EcoVLA speedups are reported on RTX 3090 with custom Triton kernels [ 12 ] , whereas SPAR uses TensorRT on Jetson AGX Orin. These speedups are not same-platform comparisons.

Fig. 3: Soft-mask budget convergence. Hidden and intermediate soft counts approach their respective 32-channel-aligned targets of 224 and 1664 (horizontal dotted lines). The vertical line marks physical pruning at step 6000.

Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Camera	Robot	Language	Light	Background	Noise	Layout	Overall ( $\boldsymbol{\Delta}$ )
$\pi_{0.5}$	baseline	100% / 100%	71.04	74.77	85.82	95.80	94.89	86.51	85.84	83.98
$\pi_{0.5}$	PAT only	90% / 10%	60.60	54.65	68.77	92.03	87.27	76.89	78.43	72.68
		70% / 10%	59.60	41.23	56.60	87.83	79.65	75.27	70.49	65.82
		50% / 10%	56.66	32.58	52.37	82.14	71.47	66.83	62.95	59.35
		30% / 10%	58.47	30.19	41.77	82.84	65.43	64.21	59.15	56.08
		10% / 10%	59.47	24.00	31.36	76.71	48.98	64.15	34.43	47.46

TABLE III: Robustness of Unpruned and Pruned

\pi_{0.5}

on LIBERO-Plus [ 17 ] . For each configuration and RL setting, we evaluate the same checkpoint on LIBERO and LIBERO-Plus. Retention targets, parameter counts, and pruning ratios follow Table II . Scores are percentages, with cross-suite results by perturbation and the aggregate under Overall. Parentheses show RL gains in percentage points. Bold highlights the best pruned score within each RL setting.

Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Pruning Ratio		Task 1 Bottle	Task 2 Cubes	Task 3 Cup	Task 4 Block	Task 5 Sorting	Task 6 Backpack	Avg. (%)
Model	Method	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Model	LLM	Task 1 Bottle	Task 2 Cubes	Task 3 Cup	Task 4 Block	Task 5 Sorting	Task 6 Backpack	Avg. (%)
$\pi_{0.5}$	baseline	100% / 100%	0%	0%	19/20	17/20	15/20	18/20	12/20	4/20	70.83
$\pi_{0.5}$	PAT	100% / 20%	40%	58%	19/20	16/20	13/20	16/20	9/20	3/20	63.33

TABLE IV: Real-Robot Task Results. We evaluate unpruned and pruned

\pi_{0.5}

on an AgileX dual-arm platform and report successes out of 20 trials per task. Target retention ratios and parameter reductions follow Table I . Avg. is the mean success rate across the six tasks described below.

Fig. 4: Real-Robot Sorting Task on the AgileX Dual-Arm Platform. Four frames from one recorded execution of Task 5: (a) initial arrangement, (b) block manipulation, (c) screw manipulation, and (d) the final recorded scene. Quantitative results are reported in Table IV .

Model	$\boldsymbol{\rho_{hid}/\rho_{int}}$	Latency (ms, $\downarrow$ )	Speedup ( $\uparrow$ )	GPU Memory (GiB, $\downarrow$ )
$\pi_{0.5}$	100% / 100%	295.04	1.00	7.098
$\pi_{0.5}$	100% / 90%	282.32	1.05	6.713
	100% / 70%	252.36	1.17	6.039
	100% / 50%	222.65	1.33	5.395
	100% / 30%	194.20	1.52	4.752
	100% / 20%	178.93	1.65	4.400

TABLE V: Inference Latency and GPU Memory of

\pi_{0.5}

. We report median full-policy latency and GPU memory using TensorRT on NVIDIA Jetson AGX Orin. Memory is in GiB, speedup is the unpruned-to-pruned latency ratio, and targets follow Table I . Best values are bold.

Method	Spatial	Object	Goal	Long	Avg.
PAT w/o $\mathcal{L}_{count}$	97.60	98.60	97.40	89.20	95.70
PAT w/o $\mathcal{L}_{entropy}$	97.40	98.40	96.60	92.00	96.10
PAT w/o $\mathcal{L}_{disc}$	97.60	98.80	97.20	90.60	96.05
PAT (full objective)	97.40	97.60	96.80	94.00	96.45

TABLE VI: Pruning-Loss Ablation for

\pi_{0.5}

, Evaluated on LIBERO. We remove one loss at a time from the full objective, targeting

\rho_{hid}=100\%

and

\rho_{int}=10\%

without RL. Scores are percentages, Long denotes LIBERO-10, and Avg. is the four-suite mean. Bold identifies the best value in each column.

Efficient VLA Model Inference

Efficient vision-language-action (VLA) model inference reduces robot-control latency, memory use, and energy while preserving task performance.

See all 190 papers

Recovering Aggressively Pruned Vision-Language-Action Models with Offline Hidden-State Distillation
Sep 17, 2026Chiyoung Kim, Sanghyuk Roy Choi, Minhyeok LeeEfficient VLA Models Vision-Language-Action Models
Chung-Ang University, Seoul, Republic of Korea
Finetuning Vision-Language-Action Models Requires Fewer Layers Than You Think
Jun 18, 2026Gia-Binh Nguyen, Trong-Bao Ho, Thien-Loc Ha +18Model Compression Efficient VLA Model Inference
Center for AI Research, VinUniversity · VinRobotics · University of Arkansas +10
Don't Run with Scissors: Pruning Breaks VLA Models but They Can Be Recovered
Oct 9, 2025Jason Jabbour, Dong-Ki Kim, Max Smith +6Vision-Language-Action Models Efficient VLA Model Inference
FieldAI · Harvard University

Sep 17, 2026J/K move · Enter open · S save

Recovering Aggressively Pruned Vision-Language-Action Models with Offline Hidden-State Distillation

Chiyoung Kim, Sanghyuk Roy Choi, Minhyeok Lee

Chung-Ang University, Seoul, Republic of Korea

Efficient VLA Models Vision-Language-Action Models VLMs for Robotics Knowledge Distillation

Open paper AskPDF arXiv

Vision-language-action (VLA) models let robots follow language instructions, but their language backbones of several billion parameters are the main obstacle to running them on robot hardware. Structured pruning reduces that backbone, and removing 63% of it from OpenVLA-OFT drops LIBERO-Long success from 93.2% to 0.8%. A recent approach restores such a model with supervised fine-tuning followed by reinforcement learning, which needs online rollouts and hundreds of GPU-hours. We recover most of the lost success entirely offline. Width pruning narrows the blocks but keeps the residual stream at its original size, so teacher and student hidden states have the same shape and are matched directly, without a projector. Training against a cache built in one teacher pass lifts the 63%-reduced student to within 3.5 points of the teacher in about 8 GPU-hours. A sweep over nine ratios locates where the recovery objective starts to matter. Up to 45% reduction the two do not differ significantly on OpenVLA-OFT. Hidden-state distillation then adds +2.1 to +4.5 points there between 63% and 87%, and +9.4 to +22.1 points on CogACT from 63% onward. At 81% on CogACT, a tripled recovery budget narrows the distilled student's gap to the teacher to 3.9 points on average, while supervised recovery stays more than 20 points below. At matched compression, width pruning yields higher success and depth pruning lower latency. On a 6-DoF manipulator, the distilled student at 72% reduction reaches 77.5% success against 59.5% for supervised recovery, runs 2.23x faster on-board than the teacher, and uses 62% less memory.

Abstract

Efficient VLA Model Inference

Explore similar work

Recovering Aggressively Pruned Vision-Language-Action Models with Offline Hidden-State Distillation

Finetuning Vision-Language-Action Models Requires Fewer Layers Than You Think

Don't Run with Scissors: Pruning Breaks VLA Models but They Can Be Recovered