KineVLA: Towards Kinematics-Aware Vision-Language-Action Models with Bi-Level Action Decomposition

• URL: http://arxiv.org/abs/2603.17524v1
• Date: Wed, 18 Mar 2026 09:28:49 GMT
• Title: KineVLA: Towards Kinematics-Aware Vision-Language-Action Models with Bi-Level Action Decomposition
• Abstract summary: We introduce a novel kinematics-rich vision-language-action (VLA) task, in which language commands densely encode diverse kinematic attributes.<n>In this setting, task goals remain invariant while execution trajectories must adapt to instruction-level kinematic specifications.<n>We show that KineVLA consistently outperforms strong VLA baselines on kinematics-sensitive benchmarks.
• Score: 89.7916653471162
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this paper, we introduce a novel kinematics-rich vision-language-action (VLA) task, in which language commands densely encode diverse kinematic attributes (such as direction, trajectory, orientation, and relative displacement) from initiation through completion, at key moments, unlike existing action instructions that capture kinematics only coarsely or partially, thereby supporting fine-grained and personalized manipulation. In this setting, where task goals remain invariant while execution trajectories must adapt to instruction-level kinematic specifications. To address this challenge, we propose KineVLA, a vision-language-action framework that explicitly decouples goal-level invariance from kinematics-level variability through a bi-level action representation and bi-level reasoning tokens to serve as explicit, supervised intermediate variables that align language and action. To support this task, we construct the kinematics-aware VLA datasets spanning both simulation and real-world robotic platforms, featuring instruction-level kinematic variations and bi-level annotations. Extensive experiments on LIBERO and a Realman-75 robot demonstrate that KineVLA consistently outperforms strong VLA baselines on kinematics-sensitive benchmarks, achieving more precise, controllable, and generalizable manipulation behaviors.

Related papers

• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data [93.7685703383343]
  Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies is surprisingly difficult.<n>We argue that this gap can be bridged gradually with the right intermediate data.<n>We show that mixing task-relevant out-of-distribution ETC data with a small amount of action data enables the model to generalize to novel visual-language conditions.
  arXiv  Detail & Related papers  (2026-06-07T08:57:51Z)
• GesVLA: Gesture-Aware Vision-Language-Action Model Embedded Representations [19.347332945780273]
  Vision-Language-Action (VLA) models have shown strong potential for general-purpose robot manipulation by unifying perception and action.<n>We introduce gesture as a parallel instruction modality and propose a Gesture-aware Vision-Language-Action model (GesVLA)<n> Experimental results show that incorporating gesture consistently improves target grounding accuracy and human-robot interaction efficiency.
  arXiv  Detail & Related papers  (2026-05-21T17:57:44Z)
• Action with Visual Primitives [36.230909065494345]
  Vision-Language-Action (VLA) models have emerged as a promising paradigm for generalist robotic manipulation.<n>We introduce AVP (Action with Visual Primitives), an end-to-end architecture that implements this visual-primitive-centric interface.<n>We show that AVP improves the success rate by 27.61% over pi_0.5 and outperforms other recent methods.
  arXiv  Detail & Related papers  (2026-05-21T08:52:47Z)
• GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization [101.37117235471709]
  We introduce GuidedVLA, a framework designed to guide the action generation to focus on task-relevant factors.<n>Our core insight is to treat the action decoder not as a monolithic learner, but as an assembly of functional components.<n>Our results suggest that explicitly guiding action-decoder learning is a promising direction for building more robust and general VLA models.
  arXiv  Detail & Related papers  (2026-05-12T16:38:40Z)
• Language-Grounded Decoupled Action Representation for Robotic Manipulation [78.42228162226839]
  We propose the Language-Grounded Decoupled Action Representation (LaDA) framework to connect perception and control.<n>LaDA introduces a fine-grained intermediate layer of three interpretable action primitives--translation, rotation, and gripper control--providing explicit semantic structure for low-level actions.<n>It further employs a semantic-guided soft-label contrastive learning objective to align similar action primitives across tasks, enhancing generalization and motion consistency.
  arXiv  Detail & Related papers  (2026-03-13T13:08:26Z)
• DAM-VLA: A Dynamic Action Model-Based Vision-Language-Action Framework for Robot Manipulation [21.418782746588263]
  DAM-VLA is a dynamic action model-based Vision-Language-Action framework.<n>It integrates VLM reasoning with diffusion-based action models for arm and gripper control.<n> DAM-VLA achieves superior success rates compared to state-of-the-art VLA methods in simulated and real-world settings.
  arXiv  Detail & Related papers  (2026-03-01T05:16:04Z)
• ActionCodec: What Makes for Good Action Tokenizers [106.78093973045526]
  Vision-Language-Action (VLA) models have demonstrated superior instruction-following and training efficiency.<n>Central to this paradigm is action tokenization, yet its design has primarily focused on reconstruction fidelity.<n>We introduce textbfActionCodec, a high-performance action tokenizer that significantly enhances both training efficiency and VLA performance.
  arXiv  Detail & Related papers  (2026-02-17T07:07:15Z)
• iFlyBot-VLA Technical Report [25.330744626382977]
  We introduce iFlyBot-VLA, a large-scale Vision-Language-Action (VLA) model trained under a novel framework.<n>The main contributions are listed as follows: (1) a latent action model thoroughly trained on large-scale human and robotic manipulation videos; (2) a dual-level action representation framework that jointly supervises both the Vision-Language Model (VLM) and the action expert during training; and (3) a mixed training strategy that combines robot trajectory data with general QA and spatial QA datasets.
  arXiv  Detail & Related papers  (2025-11-01T06:24:56Z)
• Executable Analytic Concepts as the Missing Link Between VLM Insight and Precise Manipulation [70.8381970762877]
  Vision-Language Models (VLMs) have demonstrated remarkable capabilities in semantic reasoning and task planning.<n>We introduce GRACE, a novel framework that grounds VLM-based reasoning through executable analytic concepts.<n>G GRACE provides a unified and interpretable interface between high-level instruction understanding and low-level robot control.
  arXiv  Detail & Related papers  (2025-10-09T09:08:33Z)
• Mechanistic interpretability for steering vision-language-action models [0.23371356738437823]
  Vision-Language-Action (VLA) models are a promising path to realizing generalist embodied agents.<n>We introduce the first framework for interpreting and steering VLAs via their internal representations.<n>We introduce a general-purpose activation steering method that modulates behavior in real time, without fine-tuning, reward signals, or environment interaction.
  arXiv  Detail & Related papers  (2025-08-30T03:01:57Z)
• Kinematic-aware Prompting for Generalizable Articulated Object Manipulation with LLMs [53.66070434419739]
  Generalizable articulated object manipulation is essential for home-assistant robots. We propose a kinematic-aware prompting framework that prompts Large Language Models with kinematic knowledge of objects to generate low-level motion waypoints. Our framework outperforms traditional methods on 8 categories seen and shows a powerful zero-shot capability for 8 unseen articulated object categories.
  arXiv  Detail & Related papers  (2023-11-06T03:26:41Z)
• Action Sensitivity Learning for Temporal Action Localization [35.65086250175736]
  We propose an Action Sensitivity Learning framework (ASL) to tackle the task of temporal action localization. We first introduce a lightweight Action Sensitivity Evaluator to learn the action sensitivity at the class level and instance level, respectively. Based on the action sensitivity of each frame, we design an Action Sensitive Contrastive Loss to enhance features, where the action-aware frames are sampled as positive pairs to push away the action-irrelevant frames.
  arXiv  Detail & Related papers  (2023-05-25T04:19:14Z)
• Fine-grained Temporal Contrastive Learning for Weakly-supervised Temporal Action Localization [87.47977407022492]
  This paper argues that learning by contextually comparing sequence-to-sequence distinctions offers an essential inductive bias in weakly-supervised action localization. Under a differentiable dynamic programming formulation, two complementary contrastive objectives are designed, including Fine-grained Sequence Distance (FSD) contrasting and Longest Common Subsequence (LCS) contrasting. Our method achieves state-of-the-art performance on two popular benchmarks.
  arXiv  Detail & Related papers  (2022-03-31T05:13:50Z)
• Unsupervised Motion Representation Learning with Capsule Autoencoders [54.81628825371412]
  Motion Capsule Autoencoder (MCAE) models motion in a two-level hierarchy. MCAE is evaluated on a novel Trajectory20 motion dataset and various real-world skeleton-based human action datasets.
  arXiv  Detail & Related papers  (2021-10-01T16:52:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.