Manipulation, locomotion, sim-to-real transfer and autonomous driving: learning systems that have to survive physics.
arXiv:2608.22160v1 Announce Type: new Abstract: Physical automation is scaling toward fleets of embodied machines commanded by an AI brain. Early deployments already run factories and warehouses at p...
The paper introduces CloudEdgeVLA, a cloud‑edge policy for Vision‑Language‑Action models that treats temporal misalignment as a representation‑learning problem. It encodes delayed observations into slowly varying task features on the cloud while a lightweight edge head fuses the latest cloud feature with current local vision. Experiments on four LIBERO suites show that CloudEdgeVLA retains 63.8–78.0% success under a 40‑step delay window, far outperforming VLASH and single‑path baselines.
arXiv:2608.23100v1 Announce Type: cross Abstract: Robot co-design via bi-level optimization couples within-lifetime controller learning for fitness evaluation with cross-generational morphological ev...
The paper introduces HOT-POT, a method that applies optimal transport to sparse stereo matching, addressing challenges such as occlusions, motion, and camera distortions. By modeling camera‑projected points as half‑lines and using epipolar and 3D ray distances as cost functions, the authors formulate efficient assignment problems for unsupervised sparse matching. The approach is extended to hierarchical optimal transport for unsupervised object matching, with experiments demonstrating its effectiveness in facial analysis for aligning different landmarking conventions.
ViSMoE introduces a visual‑aware sparse Mixture‑of‑Experts framework for embodied referring expression grounding, enabling an agent to navigate real environments and localize a target object from natural language instructions. By routing visual information through specialized experts, the method produces discriminative representations for both navigation views and candidate objects, unlike prior approaches that use a single vision encoder. Experiments on the REVERIE and SOON datasets show that ViSMoE surpasses existing state‑of‑the‑art methods.
The paper introduces a human‑centric video world model that can be controlled by tracked head pose and joint‑level hand poses, enabling realistic hand‑object interactions. It evaluates existing diffusion transformer conditioning methods and proposes a new 3‑D control mechanism, training a bidirectional diffusion model and distilling it into a causal, interactive system for egocentric virtual environments. Human‑subject experiments show that the system improves task performance and gives users a higher perceived sense of control compared to baseline methods.
Triplet2Track (TTS) is a closed‑loop long‑horizon imitation learning system that uses human videos to reduce robot‑collected data. It represents high‑level subgoals as instance‑grounded triplets, converts them into continuous track priors for execution, and monitors task progress from observations for online replanning. In diverse real‑world long‑horizon tasks, TTS achieves a 74.8% average success rate and supports object‑level and compositional generalization.
VeCAS is a two‑stage framework that generates contrast‑free X‑ray angiograms from non‑contrast images. Stage I localizes vascular structures using a discriminative model and cross‑modality latent distillation, while Stage II synthesizes angiographic appearance within those regions with a vessel‑focused inpainting model. Experiments on a lower‑limb vascular intervention dataset show that VeCAS improves vascular structural fidelity and image quality, and robotic guidewire navigation tests demonstrate significant reductions in time to target and operation steps.
arXiv:2608.23224v1 Announce Type: cross Abstract: Retrieval can efficiently and effectively augment a frozen vision--language--action (VLA) policy without retraining, yet retrieved text becomes a con...
Meta-Ctrl is a constrained‑decoding framework that guarantees syntactic and semantic constraints in robot plans generated by large language models. It introduces meta‑tokens—a compact vocabulary of grounded actions—to enforce syntax at the token level and semantics (preconditions, goals, ordering) at the action level, dramatically reducing memory usage from over 107 TB to under 2 GB. The approach enables a small open‑weight LM to outperform GPT‑4 on subgoal success rates in the WAH‑NL LoTa‑Bench protocol and consistently improves performance on the Embodied Agent Interface, while real‑world tabletop robot experiments show every plan satisfies its preconditions and goals by construction.
arXiv:2608.23138v1 Announce Type: cross Abstract: Vision-language-action (VLA) models often expose spatial grounding through autoregressive text coordinates or opaque action tokens, creating brittle...
arXiv:2608.22055v1 Announce Type: new Abstract: Suppose one embodied agent knows what must be built, while its teammate alone knows which transformation its workcell can perform. Neither local view d...
arXiv:2608.23238v1 Announce Type: new Abstract: We present Mover360, a controllable object manipulation framework for 360{\deg} images. Unlike perspective images, 360{\deg} images in equirectangular...
The paper introduces SASC-USOD, a framework for underwater salient object detection that learns spatially adaptive coordination between two structural representations: a boundary-sensitive representation using Laplacian filtering and a region-coherent representation via dual-range anisotropic large-kernel aggregation. A spatial coordination module estimates the relative reliability of these representations and adaptively blends them based on image content. Experiments on USOD10K and USOD benchmarks show that SASC-USOD outperforms existing methods, reducing MAE by 4.07% and 23.53% respectively, and its lightweight variant achieves 21 FPS on an NVIDIA Jetson TX2 NX.
arXiv:2605.08712v2 Announce Type: replace Abstract: Action-conditioned surgical video generation is a critical yet highly challenging problem for robotic surgery. The core difficulty is that low-dime...
The paper investigates how annotator-style rebuttals can manipulate large language model (LLM) moderation systems, either by whitewashing hateful content as normal or smearing normal content as hateful. Using a rejudge protocol that adds decision‑boundary perturbations and adversarial rationales, the authors show that such rebuttals significantly degrade moderation performance, especially in multi‑turn settings. The study finds consistent, model‑specific asymmetries between the two manipulation directions and demonstrates that explicit reasoning prompts and defensive instructions mitigate but do not eliminate the vulnerability.
arXiv:2503.09130v2 Announce Type: replace-cross Abstract: This paper presents InteractEdit, a novel framework for reference-free Human-Object Interaction (HOI) editing that tackles the challenging ta...
arXiv:2608.22799v1 Announce Type: cross Abstract: We present the NeuralNexus Arm, an open, low-cost 6-DOF robotic manipulator built by an undergraduate engineering team, together with the design deci...
arXiv:2608.21803v1 Announce Type: cross Abstract: As machine learning (ML) models are increasingly deployed in high-stakes environments, explainable AI (XAI) methods like SHAP and LIME have become es...
arXiv:2608.22971v1 Announce Type: new Abstract: Embodied Reasoning constitutes a fundamental capability of embodied intelligence, serving as the basis for autonomous perception, reasoning, and intera...