2D or 3D: Who Governs Salience in VLA Models? -- Tri-Stage Token Pruning Framework with Modality Salience Awareness

Zihao Zheng; Sicheng Tian; Zhihao Mao; Lingyue Zhang; Chenyue Li; Ziyun Zhang; Hong Gao; Yuchen Huang

cs.MMApr 10, 2026

2D or 3D: Who Governs Salience in VLA Models? -- Tri-Stage Token Pruning Framework with Modality Salience Awareness

Authors: Zihao Zheng, Sicheng Tian, Zhihao Mao, Lingyue Zhang, Chenyue Li, Ziyun Zhang, Hong Gao, Yuchen Huang, +3 more

Organizations: School of Computer Science, Peking University · School of Artificial Intelligence, Beijing Normal University · School of Computer Science, China University of Geosciences (Wuhan) · School of Electronics Engineering and Computer Science, Peking University · ZTE Corporation

Abstract

Vision-Language-Action (VLA) models have emerged as the mainstream of embodied intelligence. Recent VLA models have expanded their input modalities from 2D-only to 2D+3D paradigms, forming multi-visual-modal VLA (MVLA) models. Despite achieving improved spatial perception, MVLA faces a greater acceleration demand due to the increased number of input tokens caused by modal expansion. Token pruning is an effective optimization methods tailored to MVLA models. However, existing token pruning schemes are designed for 2D-only VLA models, ignoring 2D/3D modality salience differences. In this paper, we follow the application process of multi-modal data in MVLA models and develop a tri-stage analysis to capture the discrepancy and dynamics of 2D/3D modality salience. Based on these, we propose a corresponding tri-stage token pruning framework for MVLA models to achieve optimal 2D/3D token selection and efficient pruning. Experiments show that our framework achieves up to a 2.55x inference speedup with minimal accuracy loss, while only costing 5.8% overhead. Our Code is coming soon.

Figures & tables

Figure 1. Data Utilization Paradigm in MVLA Inference xxx

Figure 2. Comparison of Existing SVLA Token Pruning and Our Tri-Stage MVLA Token Pruning xxx

Env. &	Task	w/o Prune	Case Study 1: w/ Naive Prune						Case Study 2: w/ Naive Prune						Case Study 3: w/ Naive Prune
Model	Description	SR	SR $@$ 2D( $r$ =50%)		SR $@$ 3D( $r$ =50%)		$\mathcal{M}^{\textnormal{S1}}_{\textnormal{2D}}$	$\mathcal{M}^{\textnormal{S1}}_{\textnormal{3D}}$	SR $@$ 2D( $r$ =50%)		SR $@$ 3D( $r$ =50%)		$\mathcal{M}^{\textnormal{S1}}_{\textnormal{2D}}$	$\mathcal{M}^{\textnormal{S1}}_{\textnormal{3D}}$	SR $@$ 2D( $r$ =50%)		SR $@$ 3D( $r$ =50%)		$\mathcal{M}^{\textnormal{S1}}_{\textnormal{2D}}$	$\mathcal{M}^{\textnormal{S1}}_{\textnormal{3D}}$
MLA ( Liu et al., 2025a )	Close Box	55.00%	6.67%	$\downarrow$ 46.66%	40.00%	$\downarrow$ 13.33%	90.16%	9.84%	45.00%	$\downarrow$ 10.00%	70.00%	$\uparrow$ 23.33%	90.31%	9.69%	10.00%	$\downarrow$ 45.00%	55.00%	–	10.33%	89.67%
MLA ( Liu et al., 2025a )	Close Fridge	56.66%	33.33%	$\downarrow$ 23.33%	70.00%	$\uparrow$ 13.34%	81.47%	18.53%	50.00%	$\downarrow$ 6.66%	40.00%	$\downarrow$ 16.66%	88.24%	11.76%	5.00%	$\downarrow$ 51.66%	70%	$\uparrow$ 13.34%	85.33%	14.67%
RLBench ( James et al., 2020 )	Close Laptop	80.00%	53.33%	$\downarrow$ 13.34%	90.00%	$\uparrow$ 23.33%	90.38%	9.62%	90.00%	$\uparrow$ 13.34%	80.00%	–	91.48%	8.52%	10.00%	$\downarrow$ 70.00%	75.00%	$\downarrow$ 5.00%	90.58%	9.42%
	Swee Dustpan	66.67%	43.33%	$\downarrow$ 23.34%	96.67%	$\uparrow$ 30.00%	89.40%	10.60%	53.33%	$\downarrow$ 13.34%	70.00%	$\uparrow$ 3.33%	90.27%	9.73%	8%	$\downarrow$ 58.67%	15.33%	$\downarrow$ 51.34%	92.51%	7.49%
	Phone on Base	40.00%	35.00%	$\downarrow$ 5.00%	53.33%	$\uparrow$ 13.33%	90.43%	9.57%	33.33%	$\downarrow$ 6.67%	55.00%	$\uparrow$ 15.00%	90.12%	9.88%	5%	$\downarrow$ 35.00%	15%	$\downarrow$ 25.00%	92.63%	7.37%

Table 1. Analysis Results of Data Preprocessing Stage

Figure 3. (a) Analysis of Semantic Synthesis Stage (b) Analysis of Action Iteration Stage xxx

Figure 4. Tri-Stage Token Pruning Framework xxx

Methods	Close Box			Close Laptop			Close Fridge			Take Umbrella			Average
Methods	SR	Speed	2D/3D PR	SR	Speed	2D/3D PR	SR	Speed	2D/3D PR	SR	Speed	2D/3D PR	SR	Speed	2D/3D PR
MLA † ( Liu et al., 2025a ) w/o pruning	55.0%	1.00 $\times$	0.0% / 0.0%	80.0%	1.00 $\times$	0.0% / 0.0%	35.0%	1.00 $\times$	0.0% / 0.0%	25.0%	1.00 $\times$	0.0% / 0.0%	48.8%	1.00 $\times$	0.0% / 0.0%
Naive Prune ‡ ( $r$ =50%)	16.7%	1.22 $\times$	50.0% / 50.0%	33.3%	1.22 $\times$	50.0% / 50.0%	10.0%	1.22 $\times$	50.0% / 50.0%	6.7%	1.22 $\times$	50.0% / 50.0%	14.2%	1.22 $\times$	50.0% / 50.0%
Naive Prune ‡ ( $r$ =60%)	23.3%	1.39 $\times$	60.2% / 60.2%	46.7%	1.39 $\times$	60.2% / 60.2%	16.7%	1.39 $\times$	60.2% / 60.2%	10.0%	1.39 $\times$	60.2% / 60.2%	24.2%	1.39 $\times$	60.2% / 60.2%
Naive Prune ‡ ( $r$ =70%)	13.3%	1.47 $\times$	69.9% / 69.9%	43.3%	1.47 $\times$	69.9% / 69.9%	6.7%	1.47 $\times$	69.9% / 69.9%	3.3%	1.47 $\times$	69.9% / 69.9%	16.7%	1.47 $\times$	69.9% / 69.9%
Naive Prune ‡ ( $r$ =80%)	13.3%	1.57 $\times$	80.1% / 80.1%	30.0%	1.57 $\times$	80.1% / 80.1%	3.3%	1.57 $\times$	80.1% / 80.1%	0.0%	1.57 $\times$	80.1% / 80.1%	11.7%	1.57 $\times$	80.1% / 80.1%

Table 2. Results about Representative Tasks in RLBench Simulation Benchmark

Figure 5. Results on Simulation Benchmark xxx

Figure 6. Results on Real-World Scenarios (Task Description: Pick up the Banana and Put it on the Blue Plate) xxx

Variants	S1	S2	S3	Average
Variants	S1	S2	S3	SR	Speed
Baseline	–	–	–	70.0%	1.00 $\times$
+ Stage 1	✓	–	–	69.6%	1.22 $\times$
+ Stage 2	–	✓	–	69.8%	1.34 $\times$
+ Stage 1 + Stage 3	✓	–	✓	62.2%	1.35 $\times$
+ Stage 2 + Stage 3	–	✓	✓	71.5%	1.84 $\times$

Table 3. Results of Ablation Study

Variant	Method Cost	Latency (s)	Speedup
Baseline (MLA)	–	2.533	1.00 $\times$
+ Stage 1	12 ms	2.076	1.22 $\times$
+ Stage 2	38 ms	1.890	1.34 $\times$
+ Stage 3	11 ms	1.100	2.30 $\times$
Full (S1+S2+S3)	61 ms	1.050	2.41 $\times$

Table 4. Discussion of Overhead

Figure 8. Tri-Stage Candidate Selection Visualization (Task Description: Take Umbrella out of Umbrella Stand) xxx

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Figure 9. Common Structure of Multi-Visual-Modal VLA (MVLA) Models xxx

Explore similar work

When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA
Oct 4, 2026Tianjun Shi, Haotian Xiong, Ziyu Gong +2Efficient VLA Model Inference Vision-Language-Action Models
University of Macau
VLA-ACL: Action-Consistent Visual Token Pruning for Efficient Vision-Language-Action Models
Oct 6, 2026Owen Du, Yang Yue, Jie Zhang +3Efficient VLA Models VLMs for Robotics
ETH Zurich · Tsinghua University
SAFE-Pruner: Semantic Attention-Guided Future-Aware Token Pruning for Efficient Vision-Language-Action Manipulation
May 28, 2026Shilin Ma, Chubin Zhang, Changyuan Wang +6Efficient VLM Inference Robotic Manipulation
Tsinghua Shenzhen International Graduate School, Tsinghua University · GigaAI

Oct 4, 2026J/K move · Enter open · S save

When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA

Tianjun Shi, Haotian Xiong, Ziyu Gong, Qi Lu+1

University of Macau

Efficient VLA Model Inference Vision-Language-Action Models Visual Token Pruning

Open paper AskPDF arXiv

Visual token pruning is an effective way to accelerate vision-language models and is especially useful for vision-language-action (VLA) inference, where many visual tokens must be processed before predicting robot actions. Existing pruning methods usually estimate which tokens can be pruned based on attention scores or feature diversity, retaining tokens that are either highly attended or visually different from others. However, most of them use fixed pruning schedules, such as pruning once at a preset layer or pruning at uniformly spaced layers. Such schedules can be risky for VLA models, because the model may not know which visual regions matter for the action in early layers. Tokens that look unimportant at first may become useful after the model combines visual observations with the language instruction. In this work, we propose SAPrune, a training-free visual token pruning framework for efficient VLA inference. Instead of pruning at fixed layers, SAPrune uses a small calibration set to observe how action-to-visual attention changes across layers, and chooses pruning layers only after the attention pattern becomes more reliable. At each selected layer, SAPrune applies a dual-path pruning rule: one path protects strongly attended visual tokens from pruning, while the other prevents useful surrounding context from being discarded. Experiments on LIBERO, SIMPLER, and real-world robotic tasks show that SAPrune prunes 87.5% of visual tokens and achieves up to 1.718x inference speedup while maintaining competitive task success rates.

Abstract

Explore similar work

When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA

VLA-ACL: Action-Consistent Visual Token Pruning for Efficient Vision-Language-Action Models

SAFE-Pruner: Semantic Attention-Guided Future-Aware Token Pruning for Efficient Vision-Language-Action Manipulation