Vela: Scaling Vision-Language-Action Models with Adaptive Action Curve Parametrization

Yifan Li; Jiaxu Wang; Dongming Wu; Yicheng Jiang; Ryan Ji; Xiangyu Yue; Yanwei Fu

cs.ROOct 4, 2026

Vela: Scaling Vision-Language-Action Models with Adaptive Action Curve Parametrization

Authors: Yifan Li, Jiaxu Wang, Dongming Wu, Yicheng Jiang, Ryan Ji, Xiangyu Yue, Yanwei Fu

Organizations: Fudan University · Shanghai Innovation Institute · The Chinese University of Hong Kong · NovaXBot

Abstract

Most vision-language-action models represent future motion as fixed-rate action chunks, tying temporal resolution and prediction horizon to a fixed output budget. This pointwise representation wastes capacity on highly correlated neighboring actions, leaves temporal continuity and smoothness to be learned implicitly, and forces a tradeoff between long-horizon coverage and the local precision required for contact-rich manipulation. To address these limitations, we introduce Vela, a vision-language-action foundation model that represents future robot behavior as continuous trajectories. Vela combines a compact spline-based action representation with motion-dependent temporal support and a shared action interface for heterogeneous embodiments, allowing a fixed output budget to adapt its temporal resolution across motions. We pretrain Vela on large-scale multi-embodiment robot data and evaluate it on LIBERO-X, EBench, and two real-world long-horizon tasks, egg-cake cooking and potato shredding, obtaining promising results across simulation and physical manipulation. These results highlight the potential of continuous action representations as a foundation for future embodied foundation models. Project page and more results: https://clementine24.github.io/Vela/ .

Figures & tables

Figure 1: Overview of Vela. At inference, Vela jointly predicts a fixed set of spline control points and a motion-dependent horizon, defining a continuous action trajectory that can be sampled at arbitrary control rates. During training, motion-dependent horizon adaptation constructs supervision targets that adequately represent the demonstrated motion. The same control point budget thus covers longer spans for smooth motion and shorter spans when finer temporal resolution is needed.

Model	Level 1	Level 2	Level 3	Level 4	Level 5	Avg.
OpenVLA-OFT ( Kim et al., 2025 )	29.0	17.6	8.8	6.4	4.2	13.2
$\pi_{0}$ ( Black et al., 2024 )	29.4	21.9	11.0	7.6	5.1	15.0
X-VLA ( Zheng et al., 2026 )	30.1	22.6	10.3	6.0	4.1	14.6
OpenWAM- $\alpha$ ( Wang et al., 2026d )	35.9	28.2	21.6	13.0	10.0	21.7
GR00T N1.5 ( Bjorck et al., 2025 )	43.3	32.9	18.7	13.3	9.7	23.6
${\tau_{0}}$ -VLA ( Cai et al., 2026c )	46.5	34.2	20.5	13.0	10.5	24.9

Table 1: Evaluation on LIBERO-X ( Wang et al., 2026a ) across five progressively challenging levels. We report success rate (%), with green values indicating absolute gains over baseline

\pi_{0.5}

\pi_{0.5}

-Vela denotes variant initialized from

\pi_{0.5}

and post-trained in trajectory space.

Model	Table Top		Simple PnP		Long Horizon		Overall
Model	SR	Score	SR	Score	SR	Score	SR	Score
X-VLA ( Zheng et al., 2026 )	8.6	24	50.0	54	6.2	25	23.7	36
InternVLA-A1 ( Cai et al., 2026a )	4.3	11	43.0	47	17.9	46	23.9	36
FastWAM ( Yuan et al., 2026b )	2.9	13	49.5	53	16.9	38	25.6	37
Cosmos3-Edge ( Agarwal et al., 2026 )	18.6	32	41.5	44	24.1	49	29.3	42
GigaBrain-0.7 ( Team et al., 2026 )	37.9	59	42.0	45	20.0	37	33.3	46

Table 2: Evaluation on EBench ( Gao et al., 2026 ) across its three manipulation task families. We report success rate (SR, %) and the official task-progress Score (Score, %) with green values indicating the absolute gains over baseline

\pi_{0.5}

Figure 2: Vela executing cooking egg cake , which spans four subtasks (top to bottom).

Figure 3: Real-world results on the cooking egg cake task across four subtasks.

Configuration			Success Rate (%)
Method	Post-hoc Traj. Fit.	Max. Steps	Level 1	Level 2	Level 3	Level 4	Level 5	Avg.
	✗	10	65.2	53.2	36.0	24.1	18.0	39.3
$\pi_{0.5}$	✓	10	61.7	50.5	37.0	24.7	20.4	38.9
	✓	25	62.6	52.2	38.8	24.7	21.9	40.0
$\pi_{0.5}$ -Vela	N/A	25	66.1	53.5	40.7	27.3	22.5	42.0
Vela	N/A	25	67.2	55.9	44.9	31.5	27.1	45.3

Table 3: Ablation of post-hoc trajectory fitting (Post-hoc Traj. Fit.) on LIBERO-X. Post-hoc Traj. Fit. spline fits and resamples the predicted point-wise action chunk, whereas Vela predicts trajectory parameters directly. Max. Steps denotes the maximum number of actions executed before replanning.

Method	Level 1	Level 2	Level 3	Level 4	Level 5	Avg.
$\pi_{0.5}$	65.2	53.2	36.0	24.1	18.0	39.3
Vela w/ Fixed H.	66.4	54.3	42.8	28.1	24.4	43.2
Vela w/ Adap. H.	67.2	55.9	44.9	31.5	27.1	45.3

Table 4: Effect of adaptive temporal support on LIBERO-X. We compare Vela with globally fixed temporal support (Fixed H.) against the full model with adaptive temporal support (Adap. H.).

Objectives		Success Rate (%)
CP-FM	DT-FM	Level 1	Level 2	Level 3	Level 4	Level 5	Avg.
✓		65.2	54.7	43.1	31.3	25.8	44.0
✓	✓	65.7	56.0	44.0	31.3	27.6	44.9
	✓	67.2	55.9	44.9	31.5	27.1	45.3

Table 5: Ablation of trajectory geometry objectives on LIBERO-X.

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Task	Ours	Qwen- RobotManip	$\pi_{0.5}$	InternVLA- A1.5	$\pi_{0}$	GigaBrain- 0.7	Cosmos3- Edge	FastWAM	X-VLA
TableTop
collect coffee beans	15	5	30	0	20	50	0	0	0
flip cup collect cookies	60	90	25	30	30	40	30	0	5
frame against pen holder	65	95	45	0	25	70	20	0	30
install gear	25	50	10	15	30	20	15	15	5
peg in hole	15	10	10	0	15	15	30	0	5

Appendix

Table 6: Task-level results on EBench. Success rate (SR, %) for each of the 26 tasks, grouped into dexterous-and-precise tabletop, mobile pick-and-place (PnP), and mobile long-horizon tasks. The final row reports benchmark-wide performance.

Figure 4: Detailed capability profile of Vela and

\pi_{0.5}

on EBench. We report the official task progress Score (%) across (a) the four generalization conditions, (b) atomic skills, and (c) operating mode, temporal horizon, and precision. Vela consistently improves over

\pi_{0.5}

across the displayed categories

Figure 5: Fine-grained capability breakdown on EBench. We compare Vela with representative baselines using the official task progress Score (%) across operating mode (Mobile / Fixed-base), temporal horizon (Short / Long), precision (Low / Medium / High), and four generalization conditions (Background / Instruction / Object / Mix).

Figure 6: Robustness to replanning interval on LIBERO-X. We vary the maximum number of executed actions before replanning for both

\pi_{0.5}

and Vela.

Figure 7: Vela executing shredding potato , which spans four subtasks (top to bottom).

Figure 8: Additional real-world potato-shredding evaluation. Each bar reports the success rate over ten trials for one of the four ordered subtasks.

Explore similar work

Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models
Aug 27, 2026Senqiao Yang, Chengyao Wang, Yuxin Chen +13Diffusion-Based Vision-Language-Actions
ViDAL: A Visual Dynamics-Grounded Action Latent Space for Vision-Language-Action Models
Oct 6, 2026Yuan Xu, Yixiang Chen, Qisen Ma +10
CASIA · UCAS · Alibaba Group +1
MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action Policies
Sep 30, 2026Jingqiu Wang, Yan WangAction Prediction Weave
School of Data Science and Engineering East China Normal University, Shanghai, China

Aug 27, 2026, cs.ROJ/K move · Enter open · S save

Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models

Senqiao Yang, Chengyao Wang, Yuxin Chen, Zixuan Wang+12

Diffusion-Based Vision-Language-Actions

Open paper AskPDF arXiv

Scaling robot data is crucial for building generalist Vision-Language-Action (VLA) models, yet robot trajectories are harder to scale than web-scale image-text data because embodied collection is costly and sparsely covers the physical world. This makes representation quality a central bottleneck: under a fixed robot-data budget, continued pre-training must turn limited trajectories into transferable visual-action knowledge rather than merely fit actions. We propose VLAct, a VLA-oriented VLM backbone trained on broad, heterogeneous, multi-embodiment robot data before task-specific fine-tuning. VLAct preserves the broad VLM prior and encourages shared action semantics across embodiments through VLM-prior preservation, multi-head continuous action co-supervision, and a partially unified cross-embodiment action layout, while allowing task-specific action heads during fine-tuning. Across simulation, real-world, and unseen-embodiment transfer, VLAct consistently improves downstream performance under fixed fine-tuning protocols. On LIBERO-Plus and RoboTwin 2.0, VLAct surpasses industrial VLA systems including ABot-M0 and LingBot-VLA, achieving success rates of 82.6% and 92.5%. On RoboDojo, VLAct ranks sixth among all policies by success rate and outperforms all explicitly designated world-action model (WAM) entries on both metrics. Most notably, on RoboCasa-GR1, an unseen humanoid embodiment, VLAct using only 20% of downstream trajectories outperforms the full-data GR00T-N1.6 baseline. These results are obtained using fully open-source data and only a 16-GPU training setup, showing that representation-centric continued pre-training can deliver highly competitive performance under a modest compute budget and is an important independent axis of VLA progress beyond data scaling.

Abstract

Explore similar work

Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models

ViDAL: A Visual Dynamics-Grounded Action Latent Space for Vision-Language-Action Models

MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action Policies