Organizations: COWARobot Co. Ltd, China. · The Hong Kong Polytechnic University, Hong Kong SAR, China. · Tongji University, Shanghai, China. · Shanghai Jiao Tong University, Shanghai, China. · National University of Singapore, Singapore. · The Chinese University of Hong Kong, Hong Kong SAR, China. · Northeastern University, Shenyang, China.
Vision-Language-Action (VLA) models that directly regress joint-space actions avoid the deployment-time cost of inverse kinematics. However, their training objectives remain misaligned with task success, as joint-space losses ignore the unequal physical consequences of joint errors and visual conditioning lacks explicit focus on hand-object interaction regions. To address these issues, we propose a training-phase prior injection framework for fine-grained bimanual manipulation. First, a physics-grounded flow-matching loss maps the implicit clean-action estimate at every noise level through differentiable forward kinematics (FK) and supervises the resulting Tool Center Point (TCP) pose and inter-arm relative pose, with a bounded noise-level weighting that compensates gradient decay. Second, we introduce a global prediction query group distilled from a frozen DINOv3 teacher through future-frame feature prediction, together with a local interaction query supervised by Grounding DINO at the attention level. Both teachers are removed at inference, and the deployment pipeline is that of a standard joint-space flow policy, adding only the learned query pathway. Simulation and real-robot experiments on long-horizon bimanual tasks show that the two priors are complementary across different deployment settings. The full model excels in in-distribution deployment, while the distilled query branch is preferable under strong visual randomization.
Figures & tables
Fig. 2: The proposed loss projects intermediate clean actions into task space via differentiable FK and constrains TCP pose and inter-arm relative pose errors, providing physically supervision beyond joint-space regression.
Fig. 3: Interaction-region-guided attention distillation. Left to right: attention before distillation, Grounding DINO region targets, and attention after distillation. Local queries focus attention on task-relevant objects and grippers.
Method
Clean SR(%) ↑
Randomized SR(%) ↑
TCP err. mean(mm) ↓
TCP err. max(mm) ↓
ACT †
29.7
1.7
–
–
DP †
28.0
0.6
–
–
RDT †
34.5
13.7
–
–
π0 †
46.4
16.3
–
–
π0.5 -SFT
64.13
36.15
5.3
7.7
π0.5 -Global Query
64.26
55.29
5.1
7.2
TABLE I: Simulation results on the RoboTwin 2.0 benchmark. Success rates (SR) are averaged over 50 manipulation tasks with 100 episodes per condition. TCP err. denotes the mean and maximum deviation between the executed end-effector trajectory and the expert trajectory. † Results reported by the RoboTwin 2.0 benchmark [ 9 ] under its official protocol: Aloha AgileX embodiment, 50 clean demonstrations per task, and 100 rollouts per condition. Action Chunking with Transformers (ACT) [ 25 ] and Diffusion Policy (DP) [ 26 ] are trained from scratch; RDT [ 24 ] and π0 [ 4 ] are fine-tuned from released pretrained weights in the single-task setting.
Fig. 4: Real-world manipulation setup and results on the dual-arm PiPER-X robot. Each method is trained with 5 random seeds and each checkpoint is evaluated over 30 independent trials; bars show the mean success rate across seeds and error bars denote one standard deviation.
Vision-language-action (VLA) models show strong capabilities in single and dual-arm robotic manipulation. Prior works show coordinated bimanual behaviors can emerge from end-to-end learning, leveraging large vision-language backbones with continuous action prediction. However, as bimanual tasks become tightly coupled and execution constraints become critical, implicit coordination alone is insufficient to ensure reliable, interpretable, and stable behavior. In this work, we propose Co-VLA, a coordination-aware bimanual manipulation framework introducing explicit structural priors into VLA models. We instantiate our method on a state-of-the-art vision-language backbone by replacing its monolithic action head with a Structured Action Expert (SAE) designed for bimanual coordination. Specifically, we introduce explicit structure at the action generation level with a modular coordination-aware loss that shapes shared and residual latents according to task-specific structures. The shared latent encodes task-level coordination intent, while residual latents capture execution adjustments for each arm. At deployment, a Latent-Aware Controller (LAC) interprets the learned representations to modulate synchronization strength, execution asymmetry, smoothness, and safety constraints in real time. LAC operates at the joint-command level and remains compatible with standard control pipelines without requiring force or impedance control. Experiments across simulation and real-world benchmarks show Co-VLA significantly outperforms monolithic baselines, achieving a 27% success rate gain in tight-coordination tasks, more than doubling performance in OOD real-world scenarios (from 13% to 27%), and reducing task completion time by up to 25%.
Yandong Wang, Jiaqian Yu, Xiongfeng Peng +8
Donghua University, Shanghai, China · Samsung R&D Institute China-Beijing (SRCB), China · Samsung AI Center, DS Division, South Korea
Query-based Vision-Language-Action (VLA) models offer low-latency inference that is attractive for bimanual robotic manipulation, but we observe that they can still exhibit discontinuous actions and execution failures in complex dual-arm tasks. We hypothesize that unstable multi-view and language fusion is one contributing factor in these failures, often coinciding with attention spreading to distracting regions. To improve robustness, we introduce the Modality Masking Mechanism (M3), an embarrassingly simple, training-only strategy that requires no architectural changes or large-scale robot pretraining. M3 stochastically masks subsets of modality channels during training, exposing the policy to controlled partial observations and encouraging it to rely less on distracting cues and more on evidence that remains reliable. We evaluate M3 on ten bimanual tasks from RoboTwin 2.0 and on three long-horizon real-world tasks. Compared with the Adapter baseline, M3 improves average success by 21.7% in the Clean setting and 11.4% in Clean2Rand, where policies are trained on clean demonstrations and evaluated on randomized scenes, while also improving averaged real-world full-task success by over 30%. These results suggest that structured training-time masking is a practical way to improve the robustness of query-based VLA policies for bimanual manipulation.
Dongzhou Cheng, Ziang Li, Yixiao Zhou +6
Southeast University · Shanghai Innovation Institute · Wuhan University +4
Vision-language-action (VLA) models have shown strong potential for general-purpose robotic manipulation, but extending them to bimanual dexterous manipulation introduces a high-dimensional and structured action space. Existing bimanual VLA policies commonly concatenate right- and left-side arm--hand actions at each control timestep, leaving cross-hand interactions implicit within a single action vector. We introduce Coordi-π, a structured action representation for bimanual dexterous manipulation. Coordi-π represents the right- and left-side arm--hand actions as individual action tokens and organizes them into an interleaved sequence, explicitly exposing the two sides within the action sequence. To explicitly encode hand identity and temporal synchronization in this interleaved representation, a learnable hand-identity embedding distinguishes the two sides, while sequence--control-timestep factorized rotary positional encoding (RoPE) separates sequence position from control timestep, allowing actions at the same control timestep to share a temporal coordinate while retaining distinct sequence positions. Across all five bimanual dexterous manipulation tasks in DexJoCo, Coordi-π achieves state-of-the-art performance under both rand-obj and rand-full, improving the average success rate over π0.5 from 31.9% to 51.3% and from 18.3% to 32.3%, respectively. Coordi-π also consistently outperforms π0.5 across four real-world bimanual dexterous manipulation tasks.
Yiming Meng, Yanming Shao, Weijie Shen +4
Shanghai Jiao Tong University · The University of Hong Kong