arXiv subjects

Tongtong Cao

Publications and source records attributed to Tongtong Cao.

At least 19 recordsLinked to original sources

arXiv2026

La-JEPA: Compact Robot Policies with Latent World Modeling

World-action models adapt pretrained video generators, and vision-language-action models build on pretrained vision-language backbones; both tie policy architecture and inference cost to large transformers designed for generation rather than control. It remains unclear how much these generation-oriented choices affect control. In this work, we introduce Latent Action JEPA (La-JEPA), a compact world-action model whose policy transformer is trained from scratch in the feature space of a pretrained V-JEPA 2.1 encoder. Following Fast-WAM, La-JEPA learns to predict future features jointly with actions during training and generates only actions at inference. Among policies without embodied pretraining, La-JEPA achieves state-of-the-art performance in clean and randomized RoboTwin 2.0 scenes and on LIBERO-Plus with correct instructions, while using far fewer parameters than the strongest competing policies. It runs substantially faster than Fast-WAM at comparable performance and outperforms the closest JEPA-based methods, LeapBot-WA and JEPA-WAM, on LIBERO-Plus and RoboTwin 2.0 without their pretrained policy backbones. On four real-robot tasks, its average success is comparable to pi0.5 and above non-embodied-pretrained baselines. Through a design study of the choices behind La-JEPA, we show that JEPA features outperform video VAE latents, most clearly out of distribution, and that training-time future prediction choices significantly impact the policy even when future prediction is skipped at inference. We correct the LIBERO-Plus instructions, as the officially released evaluator corrupts them for most variants, and re-evaluate several strong open baselines. Without a generative pretrained policy backbone, La-JEPA is better and faster than far larger policies, suggesting that robot foundation models may benefit from architectures designed for control rather than inherited from generation.

cs.RO↗

arXiv2026

TurboWAM: Learning What Matters for Fast World Action Models

World action models have demonstrated promising performance in robot policy learning, yet their high inference latency remains a critical bottleneck for real-world deployment. Recent approaches have sought to improve inference efficiency by eliminating test-time future-frame imagination or accelerating action prediction. However, aggregating representations from all video world model layers can introduce redundant computation, while standard backpropagation may distribute action-supervision gradients unevenly across layers, limiting the effectiveness of world representation learning. In this work, we propose TurboWAM, a fast world action model that learns to efficiently aggregate and optimize world representations for action prediction. Specifically, TurboWAM introduces an efficient world representation aggregation mechanism that restricts aggregation to a bounded range of relevant world model layers identified through layer relevance analysis. Within this bounded range, TurboWAM estimates aggregation weights for different world model layers based on the current action denoising state and combines them with learnable layer-wise weights to fuse world representations. Moreover, TurboWAM employs a world model gradient modulation mechanism that decomposes action-supervision gradients into a layer-shared component and layer-specific residual components, enabling more balanced gradient propagation across world model layers. This mechanism operates exclusively during training and introduces no additional inference computation. Together, these designs enable more efficient world representation learning and low-latency action prediction. Extensive experiments on multiple robot manipulation benchmarks and real-world manipulation tasks show that TurboWAM achieves low inference latency while maintaining competitive task success rates compared with state-of-the-art methods.

cs.CV↗

arXiv2026

RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models

Vision-Language-Action (VLA) models remain brittle in long-horizon, contact-rich manipulation because success-only imitation provides little supervision for execution drift, while failed rollouts are often discarded. We introduce RePO-VLA, a recovery-driven policy optimization framework that assigns distinct roles to success, recovery, and failure trajectories. RePO-VLA first applies Recovery-Aware Initialization (RAI), slicing each corrective suffix into a new episode so corrective actions depend on the current adverse state rather than the preceding failure. It then learns a Progress-Aware Semantic Value Function (PAS-VF), aligning spatiotemporal trajectory features with instructions and successful references. The resulting labels salvage useful failure prefixes via reliability decay, while low-value labels mark drift and terminal breakdowns, teaching differences among nominal, failed, and corrective actions. The data engine turns adverse states into planner-generated or human-collected corrective rollouts, teaching recovery to the success manifold. Value-Conditioned Refinement (VCR) trains the policy to prefer high-progress actions. At deployment, a fixed high-value token biases actions toward the learned success manifold without online failure detectors or heuristic retries. We introduce the Failure-Recovery Benchmark (FRBench), with standardized error injection and recovery-focused evaluation. Across simulated and real-world bimanual tasks, RePO-VLA improves robustness, raising adversarial success from 20% to 75% on average and up to 80% in scaled real-world trials. Code and video results are available at https://1018weijia.github.io/RePO-VLA-Web/.

cs.RO↗

arXiv2026

Faster-WAM: Do World Action Models Need Deep Action Modules?

World Action Models (WAMs) build on pretrained video models, whose representations are grounded in physical dynamics and provide a natural basis for action prediction. Despite this natural foundation, many WAMs still rely on deep, parameter-heavy action-prediction modules that incur high inference latency and may overfit to limited robot demonstrations, restricting their real-world applicability. In this paper, we advocate a world-model-centric principle that concentrates capacity and computation in the video world model, while a lightweight action expert translates the backbone's representations into executable robot actions. We realize this principle through three key choices: Dock of Transformers (DoT) with Lite KV-Fusion to give the shallow, lightweight action expert access to representations from all video layers; world-model-only conditioning of the action expert; and retracted 1D-RoPE for positional alignment between video keys and action queries. We test this principle using Faster-WAM, a world-model-centric WAM with only a single-layer action expert. Despite this restriction on action-specific computation, Faster-WAM achieves competitive control performance on LIBERO and RoboTwin~2.0 without additional embodied pretraining. It provides approximately $3.7\times$ and $1.3\times$ inference speedups over Fast-WAM and $π_{0.5}$, respectively. Consistent with its world-model-centric design, Faster-WAM demonstrates stronger generalizability under distribution shifts: the same LIBERO-trained policy achieves $78.3\%$ success on LIBERO-Plus, exceeding Fast-WAM and LingBot-VA by $26.8$ and $8.8$ percentage points, respectively. Finally, real-robot experiments demonstrate success rates comparable to Fast-WAM, with substantially lower inference latency and shorter task-completion times.

cs.AI↗

arXiv2026

ResCue: Residual Spatial Cueing for Language-Conditioned Imitation Learning

Visual imitation policies must learn which image regions matter for action, whether applying the same operation to different objects or selecting a language-specified target. We introduce ResCue (Residual Spatial Cueing), which converts language-derived target information into a spatial condition for visual policy learning. Pretrained grounding produces a target mask; an independent cue encoder adds its features to spatial RGB features before further visual processing and pooling. This retains scene context while making the target explicit. With Diffusion Policy (DP), ResCue achieves 56.5% average success across 15 simulated tasks, versus 33.1% for DP and 40.% for the strongest baseline, $π_0$, with gains in both ordinary manipulation and instruction-dependent object selection. On three real-world bimanual tasks, it reaches 60.0%, compared with 31.7% for $π_0$. Matched-mask ablations favor residual fusion over input-level and pooled alternatives, while test-time interventions show that target-correct guidance matters for execution.

cs.RO↗

arXiv2026

H-WM: Robotic Task and Motion Planning Guided by Hierarchical World Model

World models are becoming central to robotic planning and control by predicting future state transitions. Existing approaches mainly rely on visual, latent, or language prediction, which can be difficult to ground in executable robot actions and prone to compounding errors over long horizons. In contrast, traditional robotic task and motion planning enables structured long-horizon reasoning through compact symbolic representations of world transitions, but typically lacks synchronized visual prediction. We propose Hierarchical World Model (H-WM), which jointly predicts logical and visual state transitions by combining a high-level logical world model with a low-level visual world model. The predicted logical actions and latent visual state transitions are jointly incorporated into Vision-Language-Action (VLA) models as intermediate state guidance for long-horizon task execution. Experiments on three long-horizon benchmarks and real robots show that H-WM consistently improves VLA's performance by stabilizing long-horizon execution and mitigating error accumulation. We also construct LIBERO-Logic, a frame-level aligned dataset that pairs visual observations and continuous robot states with logical actions and predicate-based logical states.

cs.RO↗

arXiv2026

RoboHarness: Memory-Driven Orchestration of Heterogeneous Robot Policies for Long-Horizon Planning

Long-horizon robotic tasks require a breadth of capabilities beyond what any single existing robot control policy can reliably provide. Combining heterogeneous policies with complementary strengths offers a promising solution, but introduces two key challenges: uncertain capability boundaries and distribution mismatches during policy handoffs. These challenges remain largely unaddressed by existing planning methods, which typically assume homogeneous, predefined skills with fixed applicability. We propose RoboHarness, a unified framework that encapsulates independently developed heterogeneous policies, including vision-language-action models (VLAs), world-action models (WAMs), reinforcement learning (RL) policies, and task and motion planners (TAMP), as reusable agentic skills. RoboHarness integrates understanding, memory, and evolution skills to reason about policy capabilities and support capability-aware task decomposition and policy routing. To mitigate distribution mismatches during policy handoffs, we introduce Memory Bridge, a plug-in policy-chaining mechanism that enables reliable transitions between heterogeneous policies without joint retraining. Extensive experiments across five public benchmarks, 500 customized tasks across 10 classes, and 135 real-robot trials demonstrate substantial gains in long-horizon and memory-dependent tasks, as well as robustness to out-of-distribution conditions.

cs.RO↗

arXiv2026

Spatial Grafting: Grounding 3D Features for Flow-Matching Robot Policies

Pretrained robot manipulation policies such as vision-language-action models (VLAs) or world-action models (WAMs) leave interaction-relevant metric geometry implicit. Recent breakthroughs in spatial reconstruction can supply the necessary geometry reliably, but their features describe local shape without stating where it lies with respect to the robot. How best to deliver these features to a pretrained policy remains unresolved. We propose Spatial Grafting, a versatile, lightweight spatial module that binds frozen reconstruction features to metric, robot-relative geometry. Spatial Grafting constructs metric-grounded spatial tokens and injects them into the flow-matching action expert through cross-attention, without modifying the host's perceptual pathway, so the host retains the full benefit of its pretraining. We evaluate it more broadly than any geometry-aware policy we compare against: one graft architecture, with no per-host redesign, on two VLAs and two WAMs, across four simulation benchmarks that span short-horizon manipulation, visual robustness, clutter and long-horizon mobile manipulation, and on three real-robot platforms with single- and dual-arm configurations. On RoboTwin 2.0, a dual-arm manipulation benchmark, the graft improves every host across VLAs and WAMs. Grafted $π_{0.5}$ gains 11.3% and 15.6% on clean and randomized scenes, reaching 94.0% and 92.4%, above the strongest published 3D-conditioned policy, WAM4D (93.8% and 89.9%). The margin widens as the horizon lengthens: on tasks from BEHAVIOR-1K, a dual-arm mobile manipulation challenge scored by average task progress, it surpasses the 2025 challenge winner on five of six tasks,by up to 0.47 Q-score, and exceeds a map-conditioned spatial policy on average across the three tasks both report.

cs.RO↗

arXiv2026

Mem2Ego: Empowering Vision-Language Models with Global-to-Ego Memory for Long-Horizon Embodied Navigation

Recent advancements in Large Language Models (LLMs) and Vision-Language Models (VLMs) have made them powerful tools in embodied navigation, enabling agents to leverage commonsense and spatial reasoning for efficient exploration in unfamiliar environments. Existing LLM-based approaches convert global memory, such as semantic or topological maps, into language descriptions to guide navigation. While this improves efficiency and reduces redundant exploration, the loss of geometric information in language-based representations hinders spatial reasoning, especially in intricate environments. To address this, VLM-based approaches directly process ego-centric visual inputs to select optimal directions for exploration. However, relying solely on a first-person perspective makes navigation a partially observed decision-making problem, leading to suboptimal decisions in complex environments. In this paper, we present a novel vision-language model (VLM)-based navigation framework that addresses these challenges by adaptively retrieving task-relevant cues from a global memory module and integrating them with the agent's egocentric observations. By dynamically aligning global contextual information with local perception, our approach enhances spatial reasoning and decision-making in long-horizon tasks. The proposed method surpasses previous state-of-the-art approaches by a significant margin on both the HSSD and HM3D benchmarks and demonstrates strong performance on a real robot.

cs.RO↗

arXiv2026

AnyBox: Efficient Zero-Shot 9DoF Pose Estimation of Boxes for Robotic Manipulation

Recovering the 9D pose of objects, both their 6D pose and 3D dimensions, under clutter and occlusion is a core requirement for warehouse automation, logistics, and manufacturing. Model-based methods are accurate but assume an instance-specific CAD model for every object, which is costly to maintain as inventories change. Model-free and category-level methods relax this assumption, yet they remain vulnerable to the symmetry, weak texture, and heavy occlusion that characterize stacked storage boxes, and they ignore the strong structural priors such scenes provide. We present \textbf{AnyBox}, an efficient zero-shot framework that exploits the geometric regularity of boxes to jointly recover pose and dimensions from a single RGB-D observation. Starting from a canonical category template, AnyBox alternates between pose and scale estimation, using the discrepancy between the reprojected template and the observed mask to drive a binary search over box dimensions. Two lightweight components make this practical: a depth-consistency filter that rejects the implausible hypotheses induced by box symmetry, and an early-stopping rule that replaces the remaining search with a single closed-form update. On public benchmarks and an in-house warehouse dataset, AnyBox improves detection AP by up to 36 points, more than doubling the previous best, and approaches instance-level pipelines that have access to ground-truth CAD models. These gains transfer downstream, raising success by 28\% on a cluttered robotic box-shelving task.

cs.CV↗

arXiv2026

Do World Action Models Generalize Better than VLAs? A Robustness Study

Robot action planning in the real world is challenging as it requires not only understanding the current state of the environment but also predicting how it will evolve in response to actions. Vision-language-action (VLA), which repurpose large-scale vision-language models for robot action generation using action experts, have achieved notable success across a variety of robotic tasks. Nevertheless, their performance remains constrained by the scope of their training data, exhibiting limited generalization to unseen scenarios and vulnerability to diverse contextual perturbations. More recently, world models have been revisited as an alternative to VLAs. These models, referred to as world action models (WAMs), are built upon world models that are trained on large corpora of video data to predict future states. With minor adaptations, their latent representation can be decoded into robot actions. It has been suggested that their explicit dynamic prediction capacity, combined with spatiotemporal priors acquired from web-scale video pretraining, enables WAMs to generalize more effectively than VLAs. In this paper, we conduct a comparative study of prominent state-of-the-art VLA policies and recently released WAMs. We evaluate their performance on the LIBERO-Plus and RoboTwin 2.0-Plus benchmarks under various visual and language perturbations. Our results show that WAMs achieve strong robustness, with LingBot-VA reaching 74.2% success rate on RoboTwin 2.0-Plus and Cosmos-Policy achieving 82.2% on LIBERO-Plus. While VLAs such as $π_{0.5}$ can achieve comparable robustness on certain tasks, they typically require extensive training with diverse robotic datasets and varied learning objectives. The evaluation code for the RoboTwin2.0-Plus benchmark is available at: https://robot-robustness.github.io/RoboTwin2.0-Plus/.

cs.RO↗

arXiv2026

Anticipate Before Acting: Future-State-Conditioned Vision-Language Navigation

End-to-end vision-language navigation (VLN) with causal vision-language models maps instructions and egocentric observations directly to actions, but standard behavior cloning supervises only the next action and does not explicitly encourage the policy state to be predictive of future visual outcomes, limiting long-horizon decision making. A privileged-input diagnostic shows that access to an expert-trajectory future image can substantially improve navigation, indicating that future observations contain rich, actionable cues, though such inputs are unavailable at deployment. Motivated by this signal, we propose Future-State-Conditioned VLN (FSC-VLN), a deployable model that augments a causal policy with a future-query token and uses training-only future-state supervision to distill information from future observations into the policy state. Concretely, during training we align the future-query representation to a frozen visual embedding $Δ$ steps ahead, while inference requires only past and current observations. This design preserves the baseline inference pattern and adds only two learned prefix tokens, implying minimal overhead. On R2R val-unseen, FSC-VLN improves SR/OSR/SPL over a StreamVLN-style baseline under two training-data regimes, with larger gains on long-horizon episodes; ablations further support the dual-query design that separates future and action queries.

cs.CV↗

arXiv2026

KAM-WM: Kinematic Affordance Maps from Latent World Models for Robot Manipulation

Learning manipulation from few demonstrations requires visual priors that capture not only where to interact, but also how the interaction should begin; static priors such as segmentation masks encode only the former. We present KAM-WM, a framework that extracts a coarse directional interaction cue from a frozen latent video world model without rollout or world-model fine-tuning. KAM-WM queries a Flow Matching image-to-video backbone once and interprets its single-step latent velocity as a Kinematic Affordance Map (KAM), which provides task-conditioned interaction regions and coarse motion structure. A lightweight Perceiver compresses KAM into tokens that condition a diffusion policy together with RGB observations and proprioception. Across LIBERO and RoboTwin2.0, KAM-WM reaches 90.6% average success on LIBERO and achieves 65.7% and 22.4% success rates in the Easy and Hard settings on RoboTwin2.0, respectively. Controlled comparisons against a zero-order mask prior suggest that part of the gains comes from directional information beyond spatial localization alone. These results indicate that, in the evaluated settings, a frozen video model can provide a useful first-order visual prior for control without the test-time cost of future rollout.

cs.RO↗

arXiv2026

Uncertainty-Aware Gaussian Map for Vision-Language Navigation

Vision-Language Navigation (VLN) requires an agent to navigate 3D environments following natural language instructions. During navigation, existing agents commonly encounter perceptual uncertainty, such as insufficient evidence for reliable grounding or ambiguity in interpreting spatial cues, yet they typically ignore such information when predicting actions. In this work, we explicitly model three forms of perceptual uncertainty (i.e., geometric, semantic, and appearance uncertainty) and integrate them into the agent's observation space to enable informed decision-making. Concretely, our agent first constructs a Semantic Gaussian Map (SGM), composed of differentiable 3D Gaussian primitives initialized from panoramic observations, that encodes both the geometric structure and semantic content of the environment. On top of SGM, geometric uncertainty is estimated through variational perturbations of Gaussian position and scale to assess structural reliability; semantic uncertainty is captured by perturbing Gaussian semantic attributes to reveal ambiguous interpretations; and appearance uncertainty is characterized by Fisher Information, which measures the sensitivity of rendered observations to Gaussian-level variations. These uncertainties are incorporated into SGM, extending it into a unified 3D Value Map, which grounds them as affordances and constraints that support reliable navigation. Comprehensive evaluations across multiple VLN benchmarks show the effectiveness of our agent.

cs.CV↗

arXiv2026

Anticipation-VLA: Solving Long-Horizon Embodied Tasks via Anticipation-based Subgoal Generation

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for embodied intelligence, enabling robots to perform tasks based on natural language instructions and current visual input. However, existing VLA models struggle with long-horizon tasks due to compounding errors. Prior methods decompose tasks into subtasks of fixed granularity, which cannot adapt to the varying complexity of execution states, limiting their robustness in long-horizon tasks. To overcome this, we introduce Anticipation Model, which adaptively and recursively generates future subgoals. This model continuously adapts as the task unfolds, adjusting future subgoals in response to evolving dynamics, facilitating more reliable planning paths. Building on this concept, we propose Anticipation-VLA, a hierarchical VLA model that leverages the anticipation model to generate actionable subgoals that guide VLA policy execution. We implement Anticipation-VLA with finetuning a Unified Multimodal Model (UMM) for high-level subgoal generation and a goal-conditioned VLA policy for low-level action execution. Experiments in both simulated and real-world robotic tasks demonstrate the effectiveness of Anticipation-VLA, highlighting the importance of adaptive and recursive subgoal generation for robust policy execution.

cs.RO↗

arXiv2026

Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning

Prevailing 2D-centric visuomotor policies exhibit a pronounced deficiency in novel view generalization, as their reliance on static observations hinders consistent action mapping across unseen views. In response, we introduce GenSplat, a feed-forward 3D Gaussian Splatting framework that facilitates view-generalized policy learning through novel view rendering. GenSplat employs a permutation-equivariant architecture to reconstruct high-fidelity 3D scenes from sparse, uncalibrated inputs in a single forward pass. To ensure structural integrity, we design a 3D-prior distillation strategy that regularizes the 3DGS optimization, preventing the geometric collapse typical of purely photometric supervision. By rendering diverse synthetic views from these stable 3D representations, we systematically augment the observational manifold during training. This augmentation forces the policy to ground its decisions in underlying 3D structures, thereby ensuring robust execution under severe spatial perturbations where baselines severely degrade.

cs.RO↗

arXiv2025

UniGaussian: Driving Scene Reconstruction from Multiple Camera Models via Unified Gaussian Representations

Urban scene reconstruction is crucial for real-world autonomous driving simulators. Although existing methods have achieved photorealistic reconstruction, they mostly focus on pinhole cameras and neglect fisheye cameras. In fact, how to effectively simulate fisheye cameras in driving scene remains an unsolved problem. In this work, we propose UniGaussian, a novel approach that learns a unified 3D Gaussian representation from multiple camera models for urban scene reconstruction in autonomous driving. Our contributions are two-fold. First, we propose a new differentiable rendering method that distorts 3D Gaussians using a series of affine transformations tailored to fisheye camera models. This addresses the compatibility issue of 3D Gaussian splatting with fisheye cameras, which is hindered by light ray distortion caused by lenses or mirrors. Besides, our method maintains real-time rendering while ensuring differentiability. Second, built on the differentiable rendering method, we design a new framework that learns a unified Gaussian representation from multiple camera models. By applying affine transformations to adapt different camera models and regularizing the shared Gaussians with supervision from different modalities, our framework learns a unified 3D Gaussian representation with input data from multiple sources and achieves holistic driving scene understanding. As a result, our approach models multiple sensors (pinhole and fisheye cameras) and modalities (depth, semantic, normal and LiDAR point clouds). Our experiments show that our method achieves superior rendering quality and fast rendering speed for driving scene simulation.

cs.CV↗

arXiv2025

CAPE: Context-Aware Diffusion Policy Via Proximal Mode Expansion for Collision Avoidance

In robotics, diffusion models can capture multi-modal trajectories from demonstrations, making them a transformative approach in imitation learning. However, achieving optimal performance following this regiment requires a large-scale dataset, which is costly to obtain, especially for challenging tasks, such as collision avoidance. In those tasks, generalization at test time demands coverage of many obstacles types and their spatial configurations, which are impractical to acquire purely via data. To remedy this problem, we propose Context-Aware diffusion policy via Proximal mode Expansion (CAPE), a framework that expands trajectory distribution modes with context-aware prior and guidance at inference via a novel prior-seeded iterative guided refinement procedure. The framework generates an initial trajectory plan and executes a short prefix trajectory, and then the remaining trajectory segment is perturbed to an intermediate noise level, forming a trajectory prior. Such a prior is context-aware and preserves task intent. Repeating the process with context-aware guided denoising iteratively expands mode support to allow finding smoother, less collision-prone trajectories. For collision avoidance, CAPE expands trajectory distribution modes with collision-aware context, enabling the sampling of collision-free trajectories in previously unseen environments while maintaining goal consistency. We evaluate CAPE on diverse manipulation tasks in cluttered unseen simulated and real-world settings and show up to 26% and 80% higher success rates respectively compared to SOTA methods, demonstrating better generalization to unseen environments.

cs.RO↗