D$^2$-VLA: Dual-Memory Dual-Frequency Vision-Language-Action Model For Long Dynamic Manipulation

Zijian Ye; Chengqi Wei; Wei Huang; Anlin Zheng; Chunyu Zou; Liangyu Wu; Zikang Zhao; Zhenjie Peng

cs.CVSep 28, 2026

D $^2$ -VLA: Dual-Memory Dual-Frequency Vision-Language-Action Model For Long Dynamic Manipulation

Authors: Zijian Ye, Chengqi Wei, Wei Huang, Anlin Zheng, Chunyu Zou, Liangyu Wu, Zikang Zhao, Zhenjie Peng, +4 more

Organizations: The University of Hong Kong · Southern University of Science and Technology

Abstract

Long-horizon manipulation requires robots to remember cues that are no longer in view while responding to moving objects. Yet vision-language-action (VLA) policies often rely on the latest observation, and refreshing their visual context typically requires another costly vision-language model (VLM) pass. We present D $^2$ -VLA, which combines dual memory and dual-frequency control at the KV-cache interface of a pretrained VLA. D $^2$ -VLA uses block-wise causal KV caching to encode observations incrementally and, guided by distinct temporal attention patterns, constructs separate historical KV read views for the VLM and action expert. Between periodic VLM updates, a gated adapter incorporates fresh visual features into the latest history-conditioned KV block, while a short fast-memory queue supports action replanning. We introduce DOMINO-Long, a ten-task benchmark requiring robots to use earlier visual cues when manipulating moving objects. D $^2$ -VLA achieves complete-task success rates of 29.3% on DOMINO, compared with 9.6% for $π_{0.5}$ and 17.2% for PUMA, and 60.0% on DOMINO-Long, compared with 35.4% and 20.6%, respectively. It improves success rates on eight real-robot tasks and reaches 97.5% on LIBERO-Long and 74.3% on RoboTwin 2.0.

Figures & tables

Figure 1: Persistent memory and responsive control with D 2 -VLA. Unlike single-frame $\pi_{0.5}$ (top), D 2 -VLA (bottom) retains temporal context through causal KV caching and incorporates fresh observations through a KV adapter for fast replanning between slow VLM updates. Real-world examples and benchmark success rates illustrate improvements in long-horizon dynamic manipulation.

Figure 2: Overview of D 2 -VLA. (a) Dual memory provides component-specific KV reads; dual-frequency control combines slow VLM refreshes with fast action replanning. (b) Block-causal attention allows bidirectional reads within each observation block. (c) The adapter refines the latest slow KV block with current visual features.

Figure 3: Attention asymmetry and token concentration. Visual attention maps (left) and token-wise attention scores (right) for the VLM (top) and action expert (bottom).

Method	SR (%) $\uparrow$	MS $\uparrow$
OpenVLA ( Kim et al., 2024 )	1.5	6.1
RDT-1B ( Liu et al., 2025 )	5.3	17.7
$\pi_{0}$ ( Black et al., 2026 )	8.2	24.0
$\pi_{0.5}$ ( Intelligence et al., 2025 )	9.6	26.2
InternVLA-M1 ( Chen et al., 2025b )	5.4	27.6
VLA-Adapter ( Wang et al., 2025 )	4.4	24.3

Table 1: Evaluation on dynamic and long-horizon manipulation. (a) Comparison on DOMINO, measured by SR and MS. (b) Per-subtask results on the 10 subtasks of DOMINO-Long.

Figure 4: Long-horizon dynamic real-world evaluation on the Unitree G1D robot. Representative task executions and success-rate comparisons with $\pi_{0.5}$ and MemoryVLA.

Method	Params (B) $\downarrow$	SR (%) $\uparrow$
Diffusion Policy ( Chi et al., 2024 )	0.1	28.0
ACT ( Zhao et al., 2023 )	0.1	29.7
DP3 ( Ze et al., 2024 )	0.3	55.2
$\pi_{0}$ ( Black et al., 2026 )	3.2	46.4
FlowPolicy ( Zhang et al., 2024 )	0.3	41.0
RDT-1B ( Liu et al., 2025 )	1.7	34.5

Table 2: Performance comparison on static manipulation benchmarks. (a) Model size and average success rate on RoboTwin 2.0, with all methods trained and evaluated exclusively on the clean setting. (b) Average success rate on LIBERO-Long.

Model	DM	DF	SR (%)
$\pi_{0.5}$	–	–	9.6
$+\,$ Causal Training	✓	–	16.1
D 2 -VLA	✓	✓	29.3

Table 3: Component ablation on DOMINO. DM: dual-memory; DF: dual-frequency.

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Consumer	Source	Current	Ret.
VLM prefill	Slow	–	0.7
Slow denoising	Slow	Yes	0.7
Fast denoising	Fast FIFO	Yes	0.7
$\rho$	0.2 (fixed)

Appendix

Table 4: Online KV selection settings for VLM prefill, slow-branch denoising, and fast-branch denoising. Source, Current, and Ret. denote historical source, current-block protection, and visual-token retention ratio, respectively.

Setting	Value
Initialization / training	$\pi_{0.5}$ Base / full fine-tuning
Slow / fast cache capacity	$4/2$ complete blocks
Action horizon $H$ / executed chunk size $E$	$25/25$
Slow-update period	$V=3E$
Main-result historical visual retention	0.7 for all active consumers
Supervised branches	3

Appendix

Table 5: Training and inference settings for DOMINO, DOMINO-Long, and the real-world tasks. Training steps count optimizer updates. Capacities include the current block.

Task	$\pi_{0.5}$		PUMA		D 2 -VLA
Task	SR $(\%)\uparrow$	MS $(\%)\uparrow$	SR $(\%)\uparrow$	MS $(\%)\uparrow$	SR $(\%)\uparrow$	MS $(\%)\uparrow$
Adjust Bottle	52.00	71.45	65.00	74.04	75.00	78.18
Beat Block Hammer	10.00	23.22	15.00	29.09	23.00	32.29
Click Alarmclock	14.00	18.47	4.00	8.38	5.00	8.74
Click Bell	0.00	6.35	3.00	13.64	11.00	19.00
Dump Bin Bigbin	0.00	16.64	0.00	39.05	38.00	44.32

Appendix

Table 6: Per-task DOMINO results under L1 + clean evaluation. Baseline SR (%) and MS for

\pi_{0.5}

and PUMA are transcribed from Table 16 of the DOMINO paper; D 2 -VLA results use a historical visual KV retention ratio of 0.7. All scores are displayed to two decimal places.

Figure 5: Overview of DOMINO-Long tasks. Visualization of the ten long-horizon manipulation tasks introduced in DOMINO-Long.

ID	Task	Information retained for later action
1	Bottle Return	Original region of the moving bottle.
2	Can Grasp	Identity of the previously observed can.
3	Stack Two	Presentation order of two blocks.
4	Two Color Cue	Colors specifying the object and destination.
5	Can Placement	Basket indicated by a transient visual cue.
6	Tray Swap	Original source region of the transferred block.

Appendix

Table 7: DOMINO-Long tasks and their memory requirements.

Task	SR	Task	SR	Task	SR
Adjust Bottle	100%	Open Microwave	86%	Place Object Stand	90%
Beat Block Hammer	82%	Pick Diverse Bottles	48%	Place Phone Stand	72%
Blocks Ranking RGB	72%	Pick Dual Bottles	62%	Place Shoe	82%
Blocks Ranking Size	64%	Place A2B Left	70%	Press Stapler	86%
Click Alarmclock	100%	Place A2B Right	72%	Put Bottles Dustbin	68%
Click Bell	100%	Place Bread Basket	58%	Put Object Cabinet	38%

Appendix

Table 8: Complete-task success rate (SR, %) on all 50 RoboTwin 2.0 Chen et al. (2025a) tasks. Task names are formatted from the corresponding environment identifiers.

Task	$\bm{\pi_{0.5}}$	M-VLA	Ours
L1: Ordered Cup Stacking	70	60	80
L2: Cross-Plate Object Swap	50	40	75
L3: Ordered Object Retrieval	45	35	65
L4: Original-Layout Restoration	40	25	65
Average	51.25	40	71.25

Appendix

Table 9: Complete-task SR (%) on the four long-horizon static real-world tasks. M-VLA and Ours denotes Memory-VLA and

\text{D}^{2}-VLA

, respectively.

Panel of Figure 8 — Figure 8: Historical KV-cache activation distributions on LIBERO and RoboTwin. In each subfigure, the left and right heatmaps show the magnitudes of cached keys (K) and values (V), respectively. The distributions are non-uniform, with stronger activations concentrated in a subset of tokens and feature dimensions. This concentration motivates selective KV-cache compression that prioritizes salient historical entries, suggesting the potential to reduce memory and computation with limited performance loss.

Figure 9: Accuracy–efficiency and historical KV-retention analysis. (a) Comparison with the single-observation baseline and Full KV Cache. LBO-L, RBT2.0, and DMO denote the LIBERO-Long Dai et al. (2026) , RoboTwin 2.0 Chen et al. (2025a) , and DOMINO Dai et al. (2026) benchmarks, respectively. SR, Lat., and Mem. refer to average success rate, inference latency, and peak GPU memory usage; (b) DOMINO success rate under different historical visual KV-retention ratios.

Setting	Period ratio $V/E$	SR (%)
Without dual-frequency mechanism	–	16.1
$V:E=50:25$	2	27.6
$V:E=75:25$	3	29.3

Appendix

Table 10: Dual-frequency ablation on DOMINO. The enabled settings use

E=25

and vary the slow-update period

V

Explore similar work

Remember What You Did: Action-History Memory with Dual-Expert Denoising for Long-Horizon Vision-Language-Action Policies
Sep 29, 2026Yaxin Zhao, Dianye Huang, Chenwei Wang +2Mamba
Harbin Institute of Technology, Harbin, China. · Medical Intelligence and Robotic Cognition (MIRoC) Lab, Department of Mechanical Engineering, The University of Hong Kong (HKU), Hong Kong SAR, China. · Department of Computing, The Hong Kong Polytechnic University, Hong Kong SAR, China
MotionVLA: Injecting Geometric Motion into Vision-Language-Action Model
Jun 6, 2026Shanglin Yuan, Weiheng Zhao, Xianda Guo +4Long-Horizon Manipulation Motion
Huazhong University of Science and Technology · D-Robotics · Wuhan University
Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation
Jul 8, 2026Hongyu Qu, Jianzhe Gao, Xiaobin Hu +6Generalist Vision--Language--Action Vision-Language-Action Framework
Nanjing University of Science and Technology · Zhejiang University · National University of Singapore

Sep 29, 2026, cs.ROJ/K move · Enter open · S save

Remember What You Did: Action-History Memory with Dual-Expert Denoising for Long-Horizon Vision-Language-Action Policies

Yaxin Zhao, Dianye Huang, Chenwei Wang, Chenguang Yang+1

Harbin Institute of Technology, Harbin, China. · Medical Intelligence and Robotic Cognition (MIRoC) Lab, Department of Mechanical Engineering, The University of Hong Kong (HKU), Hong Kong SAR, China. · Department of Computing, The Hong Kong Polytechnic University, Hong Kong SAR, China

Mamba

Open paper AskPDF arXiv

Vision-language-action (VLA) models have driven rapid progress in robotic manipulation, demonstrating strong fine-grained control and promising performance on long-horizon tasks. However, many existing VLAs lack explicit access to interaction history, making them vulnerable to perceptual aliasing: similar current observations and robot states at different task stages may induce action ambiguity and lower success rate. Existing methods incorporate temporal or progress cues through feature conditioning, action-prior modification, or sampling guidance. However, methods that jointly fine-tune memory modules and the base VLA incur additional policy-training costs, motivating the separation of trainable history-conditioned steering from frozen base-policy refinement. We propose ActMem-VLA, a dual-expert handover architecture that augments a frozen, fine-tuned VLA with a memory plugin comprising a Mamba-based memory module and a lightweight PreAction Expert (PAE). Specifically, Mamba encodes executed-action history into memory that conditions PAE alongside current context. With these inputs, PAE steers task progression during early, high-noise denoising, then passes the partially denoised action to the frozen Action Expert (AE) to refine action details during the remaining low-noise steps. The fine-tuned base VLA remains frozen throughout training, while only the Mamba module and PAE are jointly optimized. On LIBERO-Mem, ActMem-VLA achieves 80.8% average success across all ten tasks, compared with 65.2% for

π_{0.5}

and 49.5% for MemoryVLA, while introducing only 3.45% additional parameters. Across four real-world tasks, it improves the average success rate over

π_{0.5}

by 28.8%.

Abstract

Explore similar work

Remember What You Did: Action-History Memory with Dual-Expert Denoising for Long-Horizon Vision-Language-Action Policies

MotionVLA: Injecting Geometric Motion into Vision-Language-Action Model

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation