SaPaVe: Towards Active Perception and Manipulation in Vision-Language-Action Models for Robotics

• URL: http://arxiv.org/abs/2603.12193v1
• Date: Thu, 12 Mar 2026 17:23:46 GMT
• Title: SaPaVe: Towards Active Perception and Manipulation in Vision-Language-Action Models for Robotics
• Abstract summary: SaPaVe is an end-to-end framework that jointly learns active perception and manipulation capabilities.<n>We introduce ActiveViewPose-200K, a dataset of 200k image-language-camera movement pairs for semantic camera movement learning.<n>We also present ActiveManip-Bench, the first benchmark for evaluating active manipulation beyond fixed-view settings.
• Score: 45.436987571201335
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Active perception and manipulation are crucial for robots to interact with complex scenes. Existing methods struggle to unify semantic-driven active perception with robust, viewpoint-invariant execution. We propose SaPaVe, an end-to-end framework that jointly learns these capabilities in a data-efficient manner. Our approach decouples camera and manipulation actions rather than placing them in a shared action space, and follows a bottom-up training strategy: we first train semantic camera control on a large-scale dataset, then jointly optimize both action types using hybrid data. To support this framework, we introduce ActiveViewPose-200K, a dataset of 200k image-language-camera movement pairs for semantic camera movement learning, and a 3D geometry-aware module that improves execution robustness under dynamic viewpoints. We also present ActiveManip-Bench, the first benchmark for evaluating active manipulation beyond fixed-view settings. Extensive experiments in both simulation and real-world environments show that SaPaVe outperforms recent vision-language-action models such as GR00T N1 and \(π_0\), achieving up to 31.25\% higher success rates in real-world tasks. These results show that tightly coupled perception and execution, when trained with decoupled yet coordinated strategies, enable efficient and generalizable active manipulation. Project page: https://lmzpai.github.io/SaPaVe

Related papers

• 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)
• Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulation [67.16145181790522]
  This paper tackles spatial perception and manipulation challenges in Vision-Language-Action (VLA) models.<n>We propose a Geometry-Guided Gated Transformer (G3T) that aligns multi-view features under 3D geometric guidance.<n>To improve action learning efficiency, we introduce Action Manifold Learning (AML), which directly predicts actions on the valid action manifold.
  arXiv  Detail & Related papers  (2026-05-12T09:21:29Z)
• Universal Pose Pretraining for Generalizable Vision-Language-Action Policies [83.39008378156647]
  Existing Vision-Language-Action (VLA) models often suffer from feature collapse and low training efficiency.<n>We propose Pose-VLA, a decoupled paradigm that separates VLA training into a pre-training phase for extracting universal 3D spatial priors.<n>Our framework follows a two-stage pre-training pipeline, establishing fundamental spatial grounding via poses followed by motion alignment.
  arXiv  Detail & Related papers  (2026-02-23T11:00:08Z)
• MobileManiBench: Simplifying Model Verification for Mobile Manipulation [70.30578259859512]
  MobileManiBench is a large-scale benchmark for mobile-based robotic manipulation.<n>MobileManiBench features 2 mobile platforms (parallel-gripper and dexterous-hand robots), 2 synchronized cameras (head and right wrist), 630 objects in 20 categories, 5 skills (open, close, pull, push, pick) with over 100 tasks performed in 100 realistic scenes, yielding 300K trajectories.
  arXiv  Detail & Related papers  (2026-02-05T02:49:52Z)
• DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation [52.136378691610524]
  We present DynaRend, a representation learning framework that learns 3D-aware and dynamics-informed triplane features.<n>By pretraining on multi-view RGB-D video data, DynaRend jointly captures spatial geometry, future dynamics, and task semantics in a unified triplane representation.<n>We evaluate DynaRend on two challenging benchmarks, RLBench and Colosseum, demonstrating substantial improvements in policy success rate, generalization to environmental perturbations, and real-world applicability across diverse manipulation tasks.
  arXiv  Detail & Related papers  (2025-10-28T10:17:11Z)
• Precise Action-to-Video Generation Through Visual Action Prompts [62.951609704196485]
  Action-driven video generation faces a precision-generality trade-off.<n>Agent-centric action signals provide precision at the cost of cross-domain transferability.<n>We "render" actions into precise visual prompts as domain-agnostic representations.
  arXiv  Detail & Related papers  (2025-08-18T17:12:28Z)
• Polaris: Open-ended Interactive Robotic Manipulation via Syn2Real Visual Grounding and Large Language Models [53.22792173053473]
  We introduce an interactive robotic manipulation framework called Polaris. Polaris integrates perception and interaction by utilizing GPT-4 alongside grounded vision models. We propose a novel Synthetic-to-Real (Syn2Real) pose estimation pipeline.
  arXiv  Detail & Related papers  (2024-08-15T06:40:38Z)
• RoboUniView: Visual-Language Model with Unified View Representation for Robotic Manipulation [10.54770475137596]
  We propose RoboUniView, an innovative approach that decouples visual feature extraction from action learning. We first learn a unified view representation from multi-perspective views by pre-training on readily accessible data, and then derive actions from this unified view representation to control robotic manipulation. We achieve state-of-the-art performance on the demanding CALVIN benchmark, enhancing the success rate in the $D to D$ setting from 93.0% to 96.2%, and in the $ABC to D$ setting from 92.2% to 94.2%.
  arXiv  Detail & Related papers  (2024-06-27T08:13:33Z)
• Learning Manipulation by Predicting Interaction [85.57297574510507]
  We propose a general pre-training pipeline that learns Manipulation by Predicting the Interaction. The experimental results demonstrate that MPI exhibits remarkable improvement by 10% to 64% compared with previous state-of-the-art in real-world robot platforms.
  arXiv  Detail & Related papers  (2024-06-01T13:28:31Z)
• ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos [4.736059095502584]
  This work proposes a novel approach using Cross-Architecture Pseudo-Labeling with contrastive learning for semi-supervised action recognition. We introduce a novel cross-architecture approach where 3D Convolutional Neural Networks (3D CNNs) and video transformers (VIT) are utilised to capture different aspects of action representations.
  arXiv  Detail & Related papers  (2024-04-09T12:09:56Z)
• Interactive Image Segmentation with Cross-Modality Vision Transformers [18.075338835513993]
  Cross-modality vision transformers exploits mutual information to better guide the learning process. The stability of our method in term of avoiding failure cases shows its potential to be a practical annotation tool.
  arXiv  Detail & Related papers  (2023-07-05T13:29:05Z)

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.