VLMs for Robotics

VLM: Vision-Language Model

Ask about this topic

Overview

This field studies how robots can use visual observations and language instructions to understand scenes, plan tasks, and act in the physical world. Common approaches adapt pretrained vision-language models with geometric or 3D grounding, maps and affordances, action heads, skill libraries, or feedback loops; some separate slow semantic planning from fast control. Researchers train and test these systems using robot demonstrations, human videos, simulation, and real-robot tasks, with benchmarks such as LIBERO, RLBench, and Habitat. A central challenge is turning plausible language-based reasoning into reliable, metrically grounded action despite occlusion, changing scenes, latency, and safety constraints.

Latest papers

CardsList

664 papers117 papers in the last four weeks, up 10% on the four weeks before. 0.7% of all new papers.

Apr 9, 2026

RoboAgent: Chaining Basic Capabilities for Embodied Task Planning

This paper focuses on embodied task planning, where an agent acquires visual observations from the environment and executes atomic actions to accomplish a given task. Although recent Vision-Language Models (VLMs) have achieved impressive results in multimodal understanding and reasoning, their performance remains limited when applied to embodied planning that involves multi-turn interaction, long-horizon reasoning, and extended context analysis. To bridge this gap, we propose RoboAgent, a capability-driven planning pipeline in which the model actively invokes different sub-capabilities. Each capability maintains its own context, and produces intermediate reasoning results or interacts with the environment according to the query given by a scheduler. This framework decomposes complex planning into a sequence of basic vision-language problems that VLMs can better address, enabling a more transparent and controllable reasoning process. The scheduler and all capabilities are implemented with a single VLM, without relying on external tools. To train this VLM, we adopt a multi-stage paradigm that consists of: (1) behavior cloning with expert plans, (2) DAgger training using trajectories collected by the model, and (3) reinforcement learning guided by an expert policy. Across these stages, we exploit the internal information of the environment simulator to construct high-quality supervision for each capability, and we further introduce augmented and synthetic data to enhance the model's performance in more diverse scenarios. Extensive experiments on widely used embodied task planning benchmarks validate the effectiveness of the proposed approach. Our codes will be available at https://github.com/woyut/RoboAgent_CVPR26.

Peiran Xu, Jiaqi Zheng, Yadong Mu

Peking University, Beijing, China · XYZ Embodied AI, Beijing, China

Robot Task PlanningEmbodied ReasoningVLMs for Robotics
AskPDF
Apr 9, 2026

Vision-Language Navigation for Aerial Robots: Towards the Era of Large Language Models

Aerial vision-and-language navigation (Aerial VLN) aims to enable unmanned aerial vehicles (UAVs) to interpret natural language instructions and autonomously navigate complex three-dimensional environments by grounding language in visual perception. This survey provides a critical and analytical review of the Aerial VLN field, with particular attention to the recent integration of large language models (LLMs) and vision-language models (VLMs). We first formally introduce the Aerial VLN problem and define two interaction paradigms: single-instruction and dialog-based, as foundational axes. We then organize the body of Aerial VLN methods into a taxonomy of five architectural categories: sequence-to-sequence and attention-based methods, end-to-end LLM/VLM methods, hierarchical methods, multi-agent methods, and dialog-based navigation methods. For each category, we systematically analyze design rationales, technical trade-offs, and reported performance. We critically assess the evaluation infrastructure for Aerial VLN, including datasets, simulation platforms, and metrics, and identify their gaps in scale, environmental diversity, real-world grounding, and metric coverage. We consolidate cross-method comparisons on shared benchmarks and analyze key architectural trade-offs, including discrete versus continuous actions, end-to-end versus hierarchical designs, and the simulation-to-reality gap. Finally, we synthesize seven concrete open problems: long-horizon instruction grounding, viewpoint robustness, scalable spatial representation, continuous 6-DoF action execution, onboard deployment, benchmark standardization, and multi-UAV swarm navigation, with specific research directions grounded in the evidence presented throughout the survey.

Xingyu Xia, Lekai Zhou, Yujie Tang +3

Defense Innovation Institute, Chinese Academy of Military Sciences, Beijing 10071, China · Intelligent Game and Decision Laboratory, Beijing 100071, China · Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China

Vision-Language NavigationAerial VLNUAV NavigationVLMs for Robotics
AskPDF
Apr 8, 2026

HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents

We introduce HY-Embodied-0.5, a family of foundation models specifically designed for real-world embodied agents. To bridge the gap between general Vision-Language Models (VLMs) and the demands of embodied agents, our models are developed to enhance the core capabilities required by embodied intelligence: spatial and temporal visual perception, alongside advanced embodied reasoning for prediction, interaction, and planning. The HY-Embodied-0.5 suite comprises two primary variants: an efficient model with 2B activated parameters designed for edge deployment, and a powerful model with 32B activated parameters targeted for complex reasoning. To support the fine-grained visual perception essential for embodied tasks, we adopt a Mixture-of-Transformers (MoT) architecture to enable modality-specific computing. By incorporating latent tokens, this design effectively enhances the perceptual representation of the models. To improve reasoning capabilities, we introduce an iterative, self-evolving post-training paradigm. Furthermore, we employ on-policy distillation to transfer the advanced capabilities of the large model to the smaller variant, thereby maximizing the performance potential of the compact model. Extensive evaluations across 22 benchmarks, spanning visual perception, spatial reasoning, and embodied understanding, demonstrate the effectiveness of our approach. Our MoT-2B model outperforms similarly sized state-of-the-art models on 16 benchmarks, while the 32B variant achieves performance comparable to frontier models such as Gemini 3.0 Pro. In downstream robot control experiments, we leverage our robust VLM foundation to train an effective Vision-Language-Action (VLA) model, achieving compelling results in real-world physical evaluations. Code and models are open-sourced at https://github.com/Tencent-Hunyuan/HY-Embodied.

Tencent Robotics X, HY Vision Team, : +20

Mixture-of-TransformersRobot Foundation ModelsVLMs for RoboticsEmbodied ReasoningSpatial Reasoning
AskPDF
Apr 8, 2026

KITE: Keyframe-Indexed Tokenized Evidence for VLM-Based Robot Failure Analysis

We present KITE, a training-free, keyframe-anchored, layout-grounded front-end that converts long robot-execution videos into compact, interpretable tokenized evidence for vision-language models (VLMs). KITE distills each trajectory into a small set of motion-salient keyframes with open-vocabulary detections and pairs each keyframe with a schematic bird's-eye-view (BEV) representation that encodes relative object layout, axes, timestamps, and detection confidence. These visual cues are serialized with robot-profile and scene-context tokens into a unified prompt, allowing the same front-end to support failure detection, identification, localization, explanation, and correction with an off-the-shelf VLM. On the RoboFAC benchmark, KITE with Qwen2.5-VL substantially improves over vanilla Qwen2.5-VL in the training-free setting, with especially large gains on simulation failure detection, identification, and localization, while remaining competitive with a RoboFAC-tuned baseline. A small QLoRA fine-tune further improves explanation and correction quality. We also report qualitative results on real dual-arm robots, demonstrating the practical applicability of KITE as a structured and interpretable front-end for robot failure analysis. Code and models are released on our project page: https://m80hz.github.io/kite/

Mehdi Hosseinzadeh, King Hang Wong, Feras Dayoub

Australian Institute for Machine Learning (AIML), Adelaide University, Australia

Robot Failure DetectionVLMs for Robotics
AskPDF
Apr 7, 2026

ExpressMM: Expressive Mobile Manipulation Behaviors in Human-Robot Interactions

Mobile manipulators are increasingly deployed in human-centered environments to perform tasks. While completing such tasks, they should also be able to communicate their intent to the people around them using expressive robot behaviors. Prior work on expressive robot behaviors has used preprogrammed or learning-from-demonstration-based expressive motions and large language model generated high-level interactions. The majority of these existing approaches have not considered human-robot interactions (HRI) where users may interrupt, modify, or redirect a robot's actions during task execution. In this paper, we develop the novel ExpressMM framework that integrates a high-level language-guided planner based on a vision-language model for perception and conversational reasoning with a low-level vision-language-action policy to generate expressive robot behaviors during collaborative HRI tasks. Furthermore, ExpressMM supports interruptible interactions to accommodate updated or redirecting instructions by users. We demonstrate ExpressMM on a mobile manipulator assisting a human in a collaborative assembly scenario and conduct audience-based evaluation of live HRI demonstrations. Questionnaire results show that the ExpressMM-enabled expressive behaviors helped observers clearly interpret the robot's actions and intentions while supporting socially appropriate and understandable interactions. Participants also reported that the robot was useful for collaborative tasks and behaved in a predictable and safe manner during the demonstrations, fostering positive perceptions of the robot's usefulness, safety, and predictability during the collaborative tasks.

Souren Pashangpour, Haitong Wang, Matthew Lisondra +1

Autonomous Systems and Biomechatronics Laboratory (ASBLab), Department of Mechanical and Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada

Human-Robot InteractionHuman-Robot CollaborationMobile ManipulationVLMs for Robotics
AskPDF
Apr 6, 2026

ROSClaw: A Hierarchical Semantic-Physical Framework for Heterogeneous Multi-Agent Collaboration

The integration of large language models (LLMs) with embodied agents has improved high-level reasoning capabilities; however, a critical gap remains between semantic understanding and physical execution. While vision-language-action (VLA) and vision-language-navigation (VLN) systems enable robots to perform manipulation and navigation tasks from natural language instructions, they still struggle with long-horizon sequential and temporally structured tasks. Existing frameworks typically adopt modular pipelines for data collection, skill training, and policy deployment, resulting in high costs in experimental validation and policy optimization. To address these limitations, we propose ROSClaw, an agent framework for heterogeneous robots that integrates policy learning and task execution within a unified vision-language model (VLM) controller. The framework leverages e-URDF representations of heterogeneous robots as physical constraints to construct a sim-to-real topological mapping, enabling real-time access to the physical states of both simulated and real-world agents. We further incorporate a data collection and state accumulation mechanism that stores robot states, multimodal observations, and execution trajectories during real-world execution, enabling subsequent iterative policy optimization. During deployment, a unified agent maintains semantic continuity between reasoning and execution, and dynamically assigns task-specific control to different agents, thereby improving robustness in multi-policy execution. By establishing an autonomous closed-loop framework, ROSClaw minimizes the reliance on robot-specific development workflows. The framework supports hardware-level validation, automated generation of SDK-level control programs, and tool-based execution, enabling rapid cross-platform transfer and continual improvement of robotic skills. Ours project page: https://www.rosclaw.io/.

Rongfeng Zhao, Xuanhao Zhang, Zhaochen Guo +4

Shanghai Research Institute for Intelligent Autonomous Systems, Shanghai 201210, China · National Key Laboratory of Autonomous Intelligent Unmanned Systems (Tongji University), Shanghai 201210, China · Frontiers Science Center for Intelligent Autonomous Systems, Shanghai 201210, China +1

VLMs for RoboticsCross-Embodiment Robot LearningMulti-Robot CoordinationLong-Horizon Robotic ManipulationContinual Robot Learning
AskPDF
Apr 3, 2026

Learning Structured Robot Policies from Vision-Language Models via Synthetic Neuro-Symbolic Supervision

Vision-Language Models (VLMs) have recently demonstrated strong capabilities in mapping multimodal observations to robot behaviors. However, most current approaches rely on end-to-end visuomotor policies that remain opaque and difficult to analyze, limiting their use in real-world robotic applications. In contrast, classical robotic systems often rely on structured policy representations that provide interpretability, modularity, and reactive execution. This work investigates how foundation models can be specialized to generate structured robot policies grounded in multimodal perception, bridging high-dimensional learning and symbolic control. We propose a neuro-symbolic approach in which a VLM synthesizes executable Behavior Tree policies from visual observations, natural language instructions, and structured system specifications. To enable scalable supervision without manual annotation, we introduce an automated pipeline that generates a synthetic multimodal dataset of domain-randomized scenes paired with instruction-policy examples produced by a foundation model. By decoupling structured task decomposition under constrained symbolic grammars from hardware-specific motor control, we demonstrate that a 12B-parameter model can learn structured spatial-symbolic mappings required for executable BT synthesis, solely through in-silico supervision. Real-world physical experiments on two heterogeneous robotic manipulators confirm that these structurally constrained policies achieve zero-shot transfer to real-world environments. The results emphasize that the data bottleneck in robotic planning can be bypassed by procedurally synthesizing high-fidelity, neuro-symbolic training data.

Alessandro Adami, Tommaso Tubaldo, Marco Todescato +2

University of Padova, Dept. of Information Engineering, Italy. · Polytechnic of Bari Dept. of Electrical and Information Engineering, Italy. · Fraunhofer Italia Research, 39100, Bozen, Italy.

RoboticsVLMs for RoboticsBehavior TreesSynthetic Data Generation for Robotics
AskPDF
Apr 2, 2026

Tex3D: Objects as Attack Surfaces via Adversarial 3D Textures for Vision-Language-Action Models

Vision-language-action (VLA) models have shown strong performance in robotic manipulation, yet their robustness to physically realizable adversarial attacks remains underexplored. Existing studies reveal vulnerabilities through language perturbations and 2D visual attacks, but these attack surfaces are either less representative of real deployment or limited in physical realism. In contrast, adversarial 3D textures pose a more physically plausible and damaging threat, as they are naturally attached to manipulated objects and are easier to deploy in physical environments. Bringing adversarial 3D textures to VLA systems is nevertheless nontrivial. A central obstacle is that standard 3D simulators do not provide a differentiable optimization path from the VLA objective function back to object appearance, making it difficult to optimize through an end-to-end manner. To address this, we introduce Foreground-Background Decoupling (FBD), which enables differentiable texture optimization through dual-renderer alignment while preserving the original simulation environment. To further ensure that the attack remains effective across long-horizon and diverse viewpoints in the physical world, we propose Trajectory-Aware Adversarial Optimization (TAAO), which prioritizes behaviorally critical frames and stabilizes optimization with a vertex-based parameterization. Built on these designs, we present Tex3D, the first framework for end-to-end optimization of 3D adversarial textures directly within the VLA simulation environment. Experiments in both simulation and real-robot settings show that Tex3D significantly degrades VLA performance across multiple manipulation tasks, achieving task failure rates of up to 96.7%. Our empirical results expose critical vulnerabilities of VLA systems to physically grounded 3D adversarial attacks and highlight the need for robustness-aware training.

Jiawei Chen, Simin Huang, Jiawei Du +5

East China Normal University · Zhongguancun Academy · CFAR, A*STAR, Singapore +2

Adversarial Attacks on VLA ModelsAdversarial ExamplesPhysical Adversarial AttacksRobustness of VLA ModelsVLMs for Robotics
AskPDF
Apr 1, 2026

A ROS 2 Wrapper for Florence-2: Multi-Mode Local Vision-Language Inference for Robotic Systems

Foundation vision-language models are becoming increasingly relevant to robotics because they can provide richer semantic perception than narrow task-specific pipelines. However, their practical adoption in robot software stacks still depends on reproducible middleware integrations rather than on model quality alone. Florence-2 is especially attractive in this regard because it unifies captioning, optical character recognition, open-vocabulary detection, grounding and related vision-language tasks within a comparatively manageable model size. This article presents a ROS 2 wrapper for Florence-2 that exposes the model through three complementary interaction modes: continuous topic-driven processing, synchronous service calls and asynchronous actions. The wrapper is designed for local execution and supports both native installation and Docker container deployment. It also combines generic JSON outputs with standard ROS 2 message bindings for detection-oriented tasks. A functional validation is reported together with a throughput study on several GPUs, showing that local deployment is feasible with consumer grade hardware. The repository is publicly available here: https://github.com/JEDominguezVidal/florence2_ros2_wrapper

J. E. Domínguez-Vidal

VLMs for RoboticsVLM Inference
AskPDF
Mar 31, 2026

PRISM: A Multi-View Multi-Capability Retail Video Dataset for Embodied Vision-Language Models

A critical gap exists between the general-purpose visual understanding of state-of-the-art physical AI models and the specialized perceptual demands of structured real-world deployment environments. We present PRISM, a 270K-sample multi-view video supervised fine-tuning (SFT) corpus for embodied vision-language-models (VLMs) in real-world retail environments. PRISM is motivated by a simple observation - physical AI systems fail not because of poor visual recognition, but because they do not understand space, physical dynamics and embodied action well enough to operate reliably in the world. To this end, PRISM is grounded in a novel three-dimensional knowledge ontology that spans spatial knowledge, temporal and physical knowledge, and embodied action knowledge. It covers 20+ capability probes across four evaluation dimensions - Embodied Reasoning (ER), Common Sense (CS), Spatial Perception (SP), and Intuitive Physics (IP), and to our knowledge, PRISM is the first dataset to instantiate all three knowledge dimensions within a single real-world deployment domain. The corpus captures data from egocentric, exocentric and 360° viewpoints across five supermarket locations and includes open-ended, chain-of-thought, and multiple-choice supervision. At 4 fps, PRISM spans approximately 11.8M video frames and approximately 730M tokens, placing it among the largest domain-specific video SFT corpora. Fine-tuning on PRISM reduces the error rate across all 20+ probes by 66.6% over the pre-trained baseline, with significant gains in embodied action understanding where the accuracy improves by 36.4%. Our results suggest that ontology-structured, domain specific SFT can meaningfully strengthen embodied VLMs for real-world settings. The PRISM dataset and more details are available at https://dreamvu.ai/prism

Amirreza Rouhi, Parikshit Sakurikar, Satya Sai Reddy +6

DreamVu

VLMs for RoboticsVLM ReasoningReasoning for VLA Models
AskPDF
Mar 31, 2026

LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning

Existing vision-and-language navigation (VLN) models primarily reason over past and current visual observations, while largely ignoring the future visual dynamics induced by actions. As a result, they often lack an effective understanding of the causal relationship between actions and how the visual world changes, limiting robust decision-making. Humans, in contrast, can imagine the near future by leveraging action-dynamics causality, which improves both environmental understanding and navigation choices. Inspired by this capability, we propose LatentPilot, a new paradigm that exploits future observations during training as a valuable data source to learn action-conditioned visual dynamics, while requiring no access to future frames at inference. Concretely, we propose a flywheel-style training mechanism that iteratively collects on-policy trajectories and retrains the model to better match the agent's behavior distribution, with an expert takeover triggered when the agent deviates excessively. LatentPilot further learns visual latent tokens without explicit supervision; these latent tokens attend globally in a continuous latent space and are carried across steps, serving as both the current output and the next input, thereby enabling the agent to dream ahead and reason about how actions will affect subsequent observations. Experiments on R2R-CE, RxR-CE, and R2R-PE benchmarks achieve new SOTA results, and real-robot tests across diverse environments demonstrate LatentPilot's superior understanding of environment-action dynamics in scene. Project page:https://abdd.top/latentpilot/

Haihong Hao, Lei Chen, Mingfei Han +5

University of Science and Technology of China · MBZUAI · Stanford University +2

Vision-Language NavigationVLMs for RoboticsLatent Visual Reasoning
AskPDF
Mar 30, 2026

DRIVE-Nav: Directional Reasoning, Inspection, and Verification for Efficient Open-Vocabulary Navigation

Open-Vocabulary Object Navigation (OVON) requires an embodied agent to locate a language-specified target in unknown environments. Many zero-shot methods rely on frontier-candidate reasoning under incomplete observations, while topology-aware methods reduce candidate redundancy but may still introduce panoramic inspection overhead and repeated reconsideration. We present DRIVE-Nav, a structured framework that organizes exploration around persistent directions rather than raw frontiers. By inspecting encountered directions more completely and restricting subsequent decisions to still-relevant directions within a forward 240-degree view range, DRIVE-Nav reduces redundant revisits and improves path efficiency. The framework extracts and tracks directional candidates from weighted Fast Marching Method (FMM) paths, maintains representative views for semantic inspection, and combines vision-language-guided prompt enrichment with cross-frame verification to improve grounding reliability. Experiments on HM3D-OVON, HM3Dv1, HM3Dv2, and MP3D demonstrate strong overall performance and consistent efficiency gains. On HM3D-OVON, DRIVE-Nav achieves 50.2% SR and 32.6% SPL, improving the previous best method by 1.9% SR and 5.6% SPL. It also delivers the best SPL on HM3Dv1, HM3Dv2, and MP3D and transfers to a physical humanoid robot. Real-world deployment also demonstrates its effectiveness.

Maoguo Gao, Zejun Zhu, Zhiming Sun +6

1Beijing Institute of Technology · 2DeepBlue College · 3The National Key Laboratory of Autonomous Intelligent Unmanned Systems (KAIUS), School of Automation

Embodied NavigationOpen-Vocabulary Object NavigationVLMs for RoboticsObject-Goal NavigationVisual Robot Navigation
AskPDF
Mar 30, 2026

ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation

Vision-Language-Action (VLA) models and world-action models have emerged as central paradigms for general-purpose robotic intelligence, yet their empirical progress remains constrained by the absence of evaluation protocols that are both physically realistic and diagnostically controlled. Simulator-centric benchmarks provide scale and reproducibility, but cannot fully capture the reality gap induced by perception noise, contact dynamics, latency, calibration error, and hardware constraints. Conversely, real-robot evaluations are often fragmented across platforms, scenes, objects, and scoring rules, making fair comparison and failure attribution difficult. We introduce ManipArena, a standardized real-robot evaluation framework for studying manipulation generalization under matched physical conditions. ManipArena comprises 20 tasks, 10,812 expert trajectories, 13.5M frames, and approximately 188 robot hours across tabletop and mobile manipulation. The framework combines schema-defined task variation, stratified in-domain, visualshift, and semantic-OOD trials, subtask-level partial-credit scoring, three-level language annotations, low-level motor signals, and paired real-to-sim environments reconstructed from physical scenes. Using ManipArena, we evaluate seven tabletop configurations spanning VLA and world-action-model policies. The results show that real-robot conclusions depend not only on architecture, but also on model provenance, fine-tuning regime, data sampling, and annotation granularity. ManipArena thus provides a reproducible and interpretable foundation for diagnosing capability boundaries and failure modes in embodied generalization.

Yu Sun, Meng Cao, Yang Ping +24

Sun Yat-sen University · X Square Robot · MBZUAI +2

Generalization in Robotic ManipulationRobotic ManipulationRobot Manipulation BenchmarksVLMs for Robotics
AskPDF
Mar 30, 2026

Learning Multi-View Spatial Reasoning from Cross-View Relations

Vision-language models (VLMs) have achieved impressive results on single-view vision tasks, but lack the multi-view spatial reasoning capabilities essential for embodied AI systems to understand 3D environments and manipulate objects across different viewpoints. In this work, we introduce Cross-View Relations (XVR), a large-scale dataset designed to teach VLMs spatial reasoning across multiple views. XVR comprises 100K vision-question-answer samples derived from 18K diverse 3D scenes and 70K robotic manipulation trajectories, spanning three fundamental spatial reasoning tasks: Correspondence (matching objects across views), Verification (validating spatial relationships), and Localization (identifying object positions). VLMs fine-tuned on XVR achieve substantial improvements on established multi-view and robotic spatial reasoning benchmarks (MindCube and RoboSpatial). When integrated as backbones in Vision-Language-Action models, XVR-trained representations improve success rates on RoboCasa. Our results demonstrate that explicit training on cross-view spatial relations significantly enhances multi-view reasoning and transfers effectively to real-world robotic manipulation.

Suchae Jeong, Jaehwi Song, Haeone Lee +9

KAIST · Config · Hanyang University +2

Multimodal Spatial ReasoningVisual Spatial ReasoningSpatial ReasoningVLM ReasoningVLMs for Robotics
AskPDF
Mar 27, 2026

Realtime-VLA V2: Learning to Run VLAs Fast, Smooth, and Accurate

In deployment of the VLA models to real-world robotic tasks, execution speed matters. In previous work arXiv:2510.26742 we analyze how to make neural computation of VLAs on GPU fast. However, we leave the question of how to actually deploy the VLA system on the real robots open. In this report we describe a set of practical techniques to achieve the end-to-end result of running a VLA-driven robot at an impressive speed in real world tasks that require both accuracy and dexterity. The stack of technology ranges across calibration, planning & control, and learning based method to identify optimal execution speed. In the tasks we show, the robot even executes in a speed on par with casual human operation and approaching the hardware limit of our lightweight arm. The unaccelerated videos and inference traces are provided in https://dexmal.github.io/realtime-vla-v2/.

Chen Yang, Yucheng Hu, Yunchao Ma +3

Dexmal

Robotic ControlVLMs for RoboticsEfficient VLA Model Inference
AskPDF
Mar 26, 2026

Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performance

This paper proposes a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning (SFT). Some advanced finetuning methods with auxiliary training objectives can improve performance and reduce the number of convergence steps. However, they typically incur significant computational overhead due to the additional losses from auxiliary tasks. To simultaneously achieve the enhanced capabilities of auxiliary training with the simplicity of standard SFT, we decouple the two objectives of auxiliary task training within the parameter space, namely, enhancing general capabilities and fitting task-specific action distributions. To deliver this goal, we only need to train the model to converge on a small-scale task set using two distinct training strategies. The difference between the resulting model parameters can then be interpreted as capability vectors provided by auxiliary tasks. These vectors are then merged with pretrained parameters to form a capability-enhanced meta model. Moreover, when standard SFT is augmented with a lightweight orthogonal regularization loss, the merged model attains performance comparable to auxiliary finetuned baselines with reduced computational overhead. Experimental results demonstrate that this approach is highly effective across diverse robot tasks. Project page: https://chris1220313648.github.io/Fast-dVLA/

Wenxuan Song, Jiayi Chen, Shuai Chen +8

The Hong Kong University of Science and Technology (Guangzhou) · ShanghaiTech University · Shanghai Institute of Technical Physics, CAS +3

VLA Model TrainingVision-Language-Action ModelsVLMs for RoboticsModel MergingTask Vectors
AskPDF
Mar 26, 2026

ThermoAct:Thermal-Aware Vision-Language-Action Models for Robotic Perception and Decision-Making

In recent human-robot collaboration environments, there is a growing focus on integrating diverse sensor data beyond visual information to enable safer and more intelligent task execution. Although thermal data can be crucial for enhancing robot safety and operational efficiency, its integration has been relatively overlooked in prior research. This paper proposes a novel Vision-Language-Action (VLA) framework that incorporates thermal information for robot task execution. The proposed system leverages a Vision-Language Model (VLM) as a high-level planner to interpret complex natural language commands and decompose them into simpler sub-tasks. This approach facilitates efficient data collection and robust reasoning for complex operations. Unlike conventional methods that rely solely on visual data, our approach integrates thermal information, enabling the robot to perceive physical properties and proactively ensure environmental safety. Experimental results from real-world task scenarios validate the feasibility of our proposed framework, suggesting its potential to enhance task success rates and safety compared to existing vision-based systems.Project page: https://youngchae912.github.io/ThermoAct/

Young-Chae Son, Dae-Kwan Ko, Yoon-Ji Choi +1

Dongguk University

Robotic PerceptionLanguage-Conditioned Robot ControlVLMs for Robotics
AskPDF
Mar 25, 2026

3D-Mix for VLA: A Plug-and-Play Module for Integrating VGGT-based 3D Information into Vision-Language-Action Models

Vision-Language-Action (VLA) models leverage Multimodal Large Language Models (MLLMs) for robotic control, but recent studies reveal that MLLMs exhibit limited spatial intelligence due to training predominantly on 2D data, resulting in inadequate 3D perception for manipulation tasks. While recent approaches incorporate specialized 3D vision models such as VGGT to enhance spatial understanding, they employ diverse integration mechanisms without systematic investigation, leaving the optimal fusion strategy unclear. We conduct a comprehensive pilot study comparing nine VGGT integration schemes on standardized benchmarks and find that semantic-conditioned gated fusion, which adaptively balances 2D semantic and 3D geometric features based on task context, achieved the strongest performance among all nine evaluated fusion schemes in our pilot study. We present 3D-Mix, a plug-and-play module that integrates into diverse VLA architectures (GR00T-style and ππ-style) without modifying existing MLLM or action expert components. Experiments across six MLLM series (nine model variants, 2B--8B parameters) on SIMPLER and LIBERO show that 3D-Mix delivers consistent performance gains, averaging +7.0% on the out-of-domain (OOD) SIMPLER benchmark across all nine GR00T-style variants, establishing a principled approach for enhancing spatial intelligence in VLA systems.

Bin Yu, Shijie Lian, Xiaopeng Lin +8

HIT · ZGCA · HUST +5

VLMs for RoboticsVision-Language-Action Models3D VLMsRobotic Manipulation
AskPDF
Mar 25, 2026

Memory-Augmented Vision-Language Agents for Persistent and Semantically Consistent Object Captioning

Vision-Language Models (VLMs) often yield inconsistent descriptions of the same object across viewpoints, hindering the ability of embodied agents to construct consistent semantic representations over time. Previous methods resolved inconsistencies using offline multi-view aggregation or multi-stage pipelines that decouple exploration, data association, and caption learning, with limited capacity to reason over previously observed objects. In this paper, we introduce a unified, memory-augmented Vision-Language agent that simultaneously handles data association, object captioning, and exploration policy within a single autoregressive framework. The model processes the current RGB observation, a top-down explored map, and an object-level episodic memory serialized into object-level tokens, ensuring persistent object identity and semantic consistency across extended sequences. To train the model in a self-supervised manner, we collect a dataset in photorealistic 3D environments using a disagreement-based policy and a pseudo-captioning model that enforces consistency across multi-view caption histories. Extensive evaluation on a manually annotated object-level test set, demonstrate improvements of up to +11.86% in standard captioning scores and +7.39% in caption self-similarity over baseline models, while enabling scalable performance through a compact scene representation. Code, model weights, and data are available at https://hsp-iit.github.io/epos-vlm/.

Tommaso Galliena, Stefano Rosa, Tommaso Apicella +3

Italian Institute of Technology, Genoa, Italy · University of Genoa, Genoa, Italy

Memory-Augmented VLMsVLMs for RoboticsImage Captioning
AskPDF
Mar 25, 2026

ReMemNav: Memory-Based Decision Correction and Target Verification for Zero-Shot Object Navigation

Zero-shot object navigation requires agents to locate unseen targets in unfamiliar environments without prior maps or task-specific training. Despite the commonsense reasoning ability of vision-language models (VLMs), existing mapless navigators often suffer from repeated exploration and premature stopping caused by limited historical context and false-positive target predictions. We propose ReMemNav, a training-free framework that combines lightweight semantic grounding, memory-based decision correction, and target verification. RAM-derived semantic priors guide panoramic direction selection, while a geometry-triggered correction mechanism retrieves historical scene descriptions when the selected direction points toward previously visited regions. When a target is predicted to be visible, ReMemNav verifies the corresponding single-view observation before the final approach. Depth-based action sampling then converts the high-level decision into collision-free motion. Experiments on HM3D and MP3D show that ReMemNav achieves higher success rates and path efficiency than existing training-free zero-shot baselines. Specifically, ReMemNav achieves absolute SR/SPL gains of 1.7%/5.3% on HM3D v0.1, 18.2%/11.1% on HM3D v0.2, and 8.7%/6.9% on MP3D.

Feng Wu, Wei Zuo, Wenliang Yang +3

Fudan University · Tongji University

Robot NavigationVision-Language NavigationObject-Goal NavigationVLMs for Robotics
AskPDF
Mar 25, 2026

ELITE: Experiential Learning and Intent-Aware Transfer for Self-improving Embodied Agents

Vision-language models (VLMs) have shown remarkable general capabilities, yet embodied agents built on them fail at complex tasks, often skipping critical steps, proposing invalid actions, and repeating mistakes. These failures arise from a fundamental gap between the static training data of VLMs and the physical interaction for embodied tasks. VLMs can learn rich semantic knowledge from static data but lack the ability to interact with the world. To address this issue, we introduce ELITE, an embodied agent framework with {E}xperiential {L}earning and {I}ntent-aware {T}ransfer that enables agents to continuously learn from their own environment interaction experiences, and transfer acquired knowledge to procedurally similar tasks. ELITE operates through two synergistic mechanisms, i.e., self-reflective knowledge construction and intent-aware retrieval. Specifically, self-reflective knowledge construction extracts reusable strategies from execution trajectories and maintains an evolving strategy pool through structured refinement operations. Then, intent-aware retrieval identifies relevant strategies from the pool and applies them to current tasks. Experiments on the EB-ALFRED and EB-Habitat benchmarks show that ELITE achieves 9% and 5% performance improvement over base VLMs in the online setting without any supervision. In the supervised setting, ELITE generalizes effectively to unseen task categories, achieving better performance compared to state-of-the-art training-based methods. These results demonstrate the effectiveness of ELITE for bridging the gap between semantic understanding and reliable action execution.

Bingqing Wei, Zhongyu Xia, Dingai Liu +3

Wangxuan Institute of Computer Technology, Peking University, Beijing, China

Lifelong Learning AgentsVLMs for RoboticsMemory-Augmented Language Model AgentsTransfer Learning
AskPDF
Mar 24, 2026

Gaze-Regularized Vision-Language-Action Models for Robotic Manipulation

Despite advances in Vision-Language-Action (VLA) models, robotic manipulation struggles with fine-grained tasks because current models lack mechanisms for active visual attention allocation. Human gaze naturally encodes intent, planning, and execution patterns -- offering a powerful supervisory signal for guiding robot perception. We introduce a gaze-regularized training framework that aligns VLA models' internal attention with human visual patterns without architectural modifications or inference-time overhead. Our method transforms temporally aggregated gaze heatmaps into patch-level distributions and regularizes the transformer's attention through KL divergence, creating an inductive bias toward task-relevant features while preserving deployment efficiency. When integrated into existing VLA architectures, our approach yields 4-12% improvements across manipulation benchmarks. The gaze-regularized models reach equivalent performance with fewer training steps and maintain robustness under lighting variations and sensor noise. Beyond performance metrics, the learned attention patterns produce interpretable visualizations that mirror human strategies, enhancing trust in robotic systems. Moreover, our framework requires no eye-tracking equipment and applies directly to existing datasets. These results demonstrate that human perceptual priors can significantly accelerate robot learning while improving both task performance and system interpretability.

Anupam Pani, Yanchao Yang

Institute of Data Science, University Of Hong Kong

VLMs for RoboticsRobotic ManipulationLanguage-Conditioned Robot Manipulation
AskPDF
Mar 24, 2026

TRAP: Hijacking VLA CoT-Reasoning via Adversarial Patches

By integrating Chain-of-Thought (CoT) reasoning, Vision-Language-Action (VLA) models have demonstrated strong capabilities in robotic manipulation, particularly by improving generalization and interpretability. However, the security of CoT-based reasoning mechanisms remains largely unexplored. In this paper, we show that CoT reasoning introduces a novel attack vector for targeted behavior hijacking--for example, causing a robot to mistakenly deliver a knife to a person instead of an apple--without modifying the user's instruction. We first provide empirical evidence that CoT strongly governs action generation, even when it is semantically misaligned with the input instructions. Building on this observation, we propose TRAP, the first targeted behavior-hijacking adversarial attack against CoT-reasoning VLA models. By targeting the reasoning-to-action pathway, TRAP uses an adversarial patch (e.g., a tablecloth placed on the table) to steer intermediate CoT reasoning and downstream actions toward adversary-defined behaviors. Extensive evaluations on three representative reasoning VLAs, spanning distinct CoT reasoning mechanisms, demonstrate the effectiveness of TRAP. Notably, we implemented the patch by printing it on paper in a real-world setting. Our findings highlight the urgent need to secure CoT reasoning in VLA systems. The project page is available at https://zhengxian-huang.github.io/TRAP-website/.

Zhengxian Huang, Wenjun Zhu, Haoxuan Qiu +2

Zhejiang University, Hangzhou, China · Harbin Institute of Technology, Harbin, China

Adversarial Attacks on VLA ModelsRobustness of VLA ModelsVLMs for RoboticsReasoning for VLA ModelsAdversarial Patch Attacks
AskPDF
Mar 24, 2026

VLGOR: Visual-Language Knowledge Guided Offline Reinforcement Learning for Generalizable Agents

Combining Large Language Models (LLMs) with Reinforcement Learning (RL) enables agents to interpret language instructions more effectively for task execution. However, LLMs typically lack direct perception of the physical environment, which limits their understanding of environmental dynamics and their ability to generalize to unseen tasks. To address this limitation, we propose Visual-Language Knowledge-Guided Offline Reinforcement Learning (VLGOR), a framework that integrates visual and language knowledge to generate imaginary rollouts, thereby enriching the interaction data. The core premise of VLGOR is to fine-tune a vision-language model to predict future states and actions conditioned on an initial visual observation and high-level instructions, ensuring that the generated rollouts remain temporally coherent and spatially plausible. Furthermore, we employ counterfactual prompts to produce more diverse rollouts for offline RL training, enabling the agent to acquire knowledge that facilitates following language instructions while grounding in environments based on visual cues. Experiments on robotic manipulation benchmarks demonstrate that VLGOR significantly improves performance on unseen tasks requiring novel optimal policies, achieving a success rate over 24% higher than the baseline methods.

Pengsen Liu, Maosen Zeng, Nan Tang +4

National Key Laboratory for Novel Software Technology, Nanjing University, China & School of Artificial Intelligence, Nanjing University, China · Computational Intelligence Center (CIC), School of Computer and Artificial Intelligence, Shandong Jianzhu University, Jinan, 250101, China

Offline RLVLMs for RoboticsWorld Models for RoboticsRobotic Manipulation
AskPDF
Mar 24, 2026

Grounding Sim-to-Real Generalization in Robotic Manipulation: An Empirical Study with Vision-Language-Action Models

Learning a generalist control policy for robotic manipulation typically relies on large-scale datasets. Given the high cost of real-world data collection, a practical alternative is to generate synthetic data through simulation. However, the resulting synthetic data often exhibits a significant gap from real-world distributions. While many prior studies have proposed algorithms to bridge the Sim-to-Real discrepancy, there remains a lack of principled research that grounds these methods in real-world manipulation tasks, particularly their performance on generalist policies such as Vision-Language-Action (VLA) models. In this study, we empirically examine the primary determinants of Sim-to-Real generalization across four dimensions: multi-level domain randomization, photorealistic rendering, physics-realistic modeling, and reinforcement learning updates. To support this study, we design a comprehensive evaluation protocol to quantify the real-world performance of manipulation tasks. The protocol accounts for key variations in background, lighting, distractors, object types, and spatial features. Through experiments involving over 10k real-world trials, we derive critical insights into Sim-to-Real transfer. To inform and advance future studies, we release both the robotic platforms and the evaluation protocol for public access to facilitate independent verification, thereby establishing a realistic and standardized benchmark for robotic manipulation policies.

Ruixing Jin, Zicheng Zhu, Ruixiang Ouyang +4

School of Data Science, The Chinese University of Hong Kong, Shenzhen · Shenzhen Loop Area Institute

Domain RandomizationRobotics SimulationSim-to-Real TransferGeneralization in Robotic ManipulationRobotic ManipulationRobot Manipulation Benchmarks+1
AskPDF
Mar 24, 2026

CoMaTrack: Competitive Multi-Agent Game-Theoretic Tracking with Vision-Language-Action Models

Embodied Visual Tracking (EVT), a core dynamic task in embodied intelligence, requires an agent to precisely follow a language-specified target. Yet most existing methods rely on single-agent imitation learning, suffering from costly expert data and limited generalization due to static training environments. Inspired by competition-driven capability evolution, we propose CoMaTrack, a competitive game-theoretic multi-agent reinforcement learning framework that trains agents in a dynamic adversarial setting with competitive subtasks, yielding stronger adaptive planning and interference-resilient strategies. We further introduce CoMaTrack-Bench, the first open-source Habitat-based benchmark protocol and episode set for language-conditioned competitive EVT featuring dynamic dueling, featuring game scenarios between a tracker and adaptive opponents across diverse environments and instructions, enabling standardized robustness evaluation under active adversarial interactions. Experiments show that CoMaTrack achieves state-of-the-art results on both standard benchmarks and CoMaTrack-Bench. Notably, a 3B VLM trained with our framework surpasses previous single-agent imitation learning methods based on 7B models on the challenging EVT-Bench, achieving 92.1% in STT, 74.2% in DT, and 57.5% in AT. The benchmark code will be available at https://github.com/wlqcode/CoMaTrack-Bench.

Youzhi Liu, Li Gao, Liu Liu +2

Amap, Alibaba Group

Game-Theoretic RLReinforcement LearningVLMs for RoboticsRobustness of VLA ModelsSequential MARL
AskPDF
Mar 24, 2026

CATNAV: Cached Vision-Language Traversability for Efficient Zero-Shot Robot Navigation

Navigating unstructured environments requires assessing traversal risk relative to a robot's physical capabilities, a challenge that varies across embodiments. We present CATNAV, a cost-aware traversability navigation framework that leverages multimodal LLMs for zero-shot, embodiment-aware costmap generation without task-specific training. We introduce a visuosemantic caching mechanism that detects scene novelty and reuses prior risk assessments for semantically similar frames, reducing online VLM queries by 85.7%. Furthermore, we introduce a VLM-based trajectory selection module that evaluates proposals through visual reasoning to choose the safest path given behavioral constraints. We evaluate CATNAV on a quadruped robot across indoor and outdoor unstructured environments, comparing against state-of-the-art vision-language-action baselines. Across five navigation tasks, CATNAV achieves 10 percentage point higher average goal-reaching rate and 33% fewer behavioral constraint violations.

Aditya Potnis, Francisco Affonso, Shreya Gummadi +2

Field Robotics Engineering and Science Hub (FRESH), Illinois Autonomous Farm, University of Illinois at Urbana-Champaign (UIUC), IL

Robot Traversability EstimationRobot NavigationVisual Robot NavigationVLMs for RoboticsEfficient VLM Inference
AskPDF
Mar 23, 2026

ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model

Recent progress in latent world models (e.g., V-JEPA2) has shown promising capability in forecasting future world states from video observations. Nevertheless, dense prediction from a short observation window limits temporal context and can bias predictors toward local, low-level extrapolation, making it difficult to capture long-horizon semantics and reducing downstream utility. Vision--language models (VLMs), in contrast, provide strong semantic grounding and general knowledge by reasoning over uniformly sampled frames, but they are not ideal as standalone dense predictors due to compute-driven sparse sampling, a language-output bottleneck that compresses fine-grained interaction states into text-oriented representations, and a data-regime mismatch when adapting to small action-conditioned datasets. We propose a VLM-guided JEPA-style latent world modeling framework that combines dense-frame dynamics modeling with long-horizon semantic guidance via a dual-temporal pathway: a dense JEPA branch for fine-grained motion and interaction cues, and a uniformly sampled VLM thinker branch with a larger temporal stride for knowledge-rich guidance. To transfer the VLM's progressive reasoning signals effectively, we introduce a hierarchical pyramid representation extraction module that aggregates multi-layer VLM representations into guidance features compatible with latent prediction. Experiments on hand-manipulation trajectory prediction show that our method outperforms both a strong VLM-only baseline and a JEPA-predictor baseline, and yields more robust long-horizon rollout behavior.

Haichao Zhang, Yijiang Li, Shwai He +5

Northeastern University · University of California San Diego · University of Maryland +2

Latent World ModelsTrajectory PredictionJoint-Embedding Predictive ArchitectureVLMs for Robotics
AskPDF
Mar 23, 2026

DualCoT-VLA: Visual-Linguistic Chain of Thought via Parallel Reasoning for Vision-Language-Action Models

Vision-Language-Action (VLA) models map visual observations and language instructions directly to robotic actions. While effective for simple tasks, standard VLA models often struggle with complex, multi-step tasks requiring logical planning, as well as precise manipulations demanding fine-grained spatial perception. Recent efforts have incorporated Chain-of-Thought (CoT) reasoning to endow VLA models with a ``thinking before acting'' capability. However, current CoT-based VLA models face two critical limitations: 1) an inability to simultaneously capture low-level visual details and high-level logical planning due to their reliance on isolated, single-modal CoT; 2) high inference latency with compounding errors caused by step-by-step autoregressive decoding. To address these limitations, we propose DualCoT-VLA, a visual-linguistic CoT method for VLA models with a parallel reasoning mechanism. To achieve comprehensive multi-modal reasoning, our method integrates a visual CoT for low-level spatial understanding and a linguistic CoT for high-level task planning. Furthermore, to overcome the latency bottleneck, we introduce a parallel CoT mechanism that incorporates two sets of learnable query tokens, shifting autoregressive reasoning to single-step forward reasoning. Extensive experiments demonstrate that our DualCoT-VLA achieves state-of-the-art performance on the LIBERO and RoboCasa GR1 benchmarks, as well as in real-world platforms.

Zhide Zhong, Junfeng Li, Junjie He +10

The Hong Kong University of Science and Technology (Guangzhou) · Huawei Foundation Model Department

VLMs for RoboticsReasoning for VLA ModelsEfficient VLA Model Inference
AskPDF
Mar 23, 2026

UniDex: A Robot Foundation Suite for Universal Dexterous Hand Control from Egocentric Human Videos

Dexterous manipulation remains challenging due to the cost of collecting real-robot teleoperation data, the heterogeneity of hand embodiments, and the high dimensionality of control. We present UniDex, a robot foundation suite that couples a large-scale robot-centric dataset with a unified vision-language-action (VLA) policy and a practical human-data capture setup for universal dexterous hand control. First, we construct UniDex-Dataset, a robot-centric dataset over 50K trajectories across eight dexterous hands (6--24 DoFs), derived from egocentric human video datasets. To transform human data into robot-executable trajectories, we employ a human-in-the-loop retargeting procedure to align fingertip trajectories while preserving plausible hand-object contacts, and we operate on explicit 3D pointclouds with human hands masked to narrow kinematic and visual gaps. Second, we introduce the Function-Actuator-Aligned Space (FAAS), a unified action space that maps functionally similar actuators to shared coordinates, enabling cross-hand transfer. Leveraging FAAS as the action parameterization, we train UniDex-VLA, a 3D VLA policy pretrained on UniDex-Dataset and finetuned with task demonstrations. In addition, we build UniDex-Cap, a simple portable capture setup that records synchronized RGB-D streams and human hand poses and converts them into robot-executable trajectories to enable human-robot data co-training that reduces reliance on costly robot demonstrations. On challenging tool-use tasks across two different hands, UniDex-VLA achieves 81% average task progress and outperforms prior VLA baselines by a large margin, while exhibiting strong spatial, object, and zero-shot cross-hand generalization. Together, UniDex-Dataset, UniDex-VLA, and UniDex-Cap provide a scalable foundation suite for universal dexterous manipulation.

Gu Zhang, Qicheng Xu, Haozhe Zhang +16

Tsinghua University · Shanghai Qizhi Institute · Sun Yat-sen University +1

Robotic ManipulationCross-Embodiment Robot LearningRobot Imitation LearningVLMs for RoboticsRobotic Data Collection
AskPDF