Manipulation, locomotion, sim-to-real transfer and autonomous driving: learning systems that have to survive physics.
We present ParticleSplat, a self-supervised object-centric learning method that decomposes scenes into a set of latent ''particles'' representing semantic entities through feedforward 3D Gaussian Spla...
World models endow perceptual systems with the ability to predict how scenes evolve under interaction. They are most beneficial when trained on diverse volumes of data, to instill a rich prior into do...
Recent advances in video generation allow robots to learn manipulation trajectories from generated videos. However, these approaches produce purely kinematic trajectories that lack force information,...
Factory work is a promising early scenario for embodied AI: assigning repetitive manual jobs to robots has clear economic payoff, and a structured station keeps the jobs tractable for current policies...
Self-driving laboratories can explore synthesis conditions autonomously, but their decision-making layer is typically a black-box optimizer, and the output is a set of optimized samples, with the meas...
Collective intelligence is a collaborative autonomy paradigm in which multiple agents pursue shared objectives through local perception, information exchange, and coordinated action. UAV swarms embody...
Vision-Language-Action models and World-Action Models have advanced language-conditioned robotic manipulation, yet often leave metric relations among actions, objects, and scene geometry implicit. Hum...
This paper studies energy-aware manipulation as a physically grounded learning problem. We define a joint-space mechanical-work proxy from joint torque and angular displacement, and train a differenti...
Cross-view object geo-localization (CVOGL) locates a target in satellite imagery using drone or street-view queries. Existing methods train separate detectors for each viewpoint, leading to parameter...
The integration of large-scale foundation models with physical embodiments has led to significant advancements in robotics known as Generative Physical Artificial Intelligence (GPAI). These agentic AI...
The paper investigates how different 2D image backbones affect indoor semantic occupancy prediction in RGB‑D pipelines. By keeping the projection, depth branch, and occupancy head constant and swapping only the image encoder, the authors find that stronger backbones such as DINOv2 and BLIP2 significantly raise mIoU scores compared to the default ResNet‑50. These results show that the choice of image backbone is a major determinant of 3D occupancy accuracy, outweighing many specialized 3D modules.
The paper introduces FastFlowGS, a streaming 4D Gaussian Splatting method that reconstructs fast-moving subjects from a sparse set of external cameras, and Monaco4D, a photorealistic Unreal Engine 5 benchmark featuring Formula 1 sequences with dense ground truth. FastFlowGS combines sparse matches, semi-dense tracks, and dense optical flow using a Kalman-style temporal update, achieving significant performance gains over existing baselines on both CMU-Panoptic and Monaco4D datasets. The benchmark provides varied illumination and viewpoints from trackside, onboard, and drone cameras, enabling evaluation of high-speed outdoor reconstruction.
EMODY Flow is a lightweight flow‑matching framework that generates synchronized full‑body motion and facial expressions conditioned on speech and emotion. It attaches to a frozen Qwen‑3 Omni model, reusing its audio codecs to drive two parallel DiT generators for SMPL‑X body pose and FLAME facial expressions. An auxiliary emotion classifier at training time restores emotion sensitivity, enabling EMODY Flow to achieve state‑of‑the‑art gesture quality on BEAT2 and zero‑shot facial animation on TFHP, with significant improvements in FGD, Beat Correlation, and Diversity metrics.
Det‑LIME is a detector‑aware, multi‑instance adaptation of LIME designed to explain black‑box object detectors used in marine mammal research. It generates instance‑specific, box‑aligned explanations by weighting detections, applying a proximity kernel, and using IoU‑based matching to track instances across perturbations. Evaluated on aerial drone imagery of harbor seals and a seabird case study, Det‑LIME outperformed vanilla LIME, Stabilized LIME, Deterministic LIME, and gradient‑based methods in Attribution Ratio and Max Saliency Hit Rate, offering higher‑resolution, instance‑aware explanations that aid debugging, data augmentation, and modeling improvements.
The paper re‑examines the impact of removing the encoder from Action Chunking Transformers (ACT), a model used for robot manipulation learning. Contrary to the original claim that encoder removal drops success rates from 35% to 2%, the authors find no such dramatic effect in their re‑runs, though minor variations remain uncertain. They attribute discrepancies to training length and checkpoint selection, and note that the encoder’s latent variable offers little reconstruction benefit on the tested benchmark, while its removal speeds up training.
arXiv:2512.15708v2 Announce Type: replace Abstract: Foundation models are vital tools in various Computer Vision applications. They take as input a single RGB image and output a deep feature represen...
TARC (Time‑Adaptive Robotic Control) is a reinforcement‑learning framework that lets a policy predict both a control action and how long it should be applied, thereby learning temporally extended actions. By optimizing task performance under constraints on the number of control switches, TARC can adapt its control rate online, using high‑frequency feedback only when necessary. Experiments on a high‑speed RC car, a Unitree Go1 quadruped, and a vision‑language action model show that TARC matches the performance of high‑frequency discrete‑time controllers while operating at less than half their control frequency.
Mem-World introduces a memory‑augmented action‑conditioned world model for robot manipulation, featuring W‑VMem—a 4D wrist‑view‑centered surfel‑indexed memory that anchors historical observations to evolving surface elements. By explicitly modeling when and where scene elements are observed, the system retrieves geometry‑aware history frames during generation, providing informative, non‑redundant context for future action predictions. Experiments demonstrate that Mem‑World produces persistent rollouts, improves policy evaluation reliability (14.5 % higher Pearson correlation with real‑world performance), and boosts long‑horizon task success rates from 58 % to 72 % using synthetic data generation.
SlotDiT introduces a text-guided Diffusion Transformer that operates in a slot-based latent space, decomposing scenes into object-centric slots and autoregressively denoising future slot trajectories to predict scene dynamics. The model is conditioned on a reference image and a language instruction, enabling it to generate video content that reflects both visual context and textual guidance. Experiments comparing slot-based representations to VAE-based and semantics-aligned alternatives show that SlotDiT achieves competitive video generation quality while improving task-completion rates across four robotic datasets and offering a more computationally efficient latent representation.
The survey "World-Action Models for Robot Learning and Control" reviews recent advances in coupling future world prediction with executable action generation for robots in open environments. It clarifies the scope of World-Action Models (WAMs) relative to conventional world models, model-based RL, and Vision‑Language‑Action policies, and organizes existing methods through a unified taxonomy covering representations, transition modeling, action interfaces, architectures, training pipelines, data modalities, and scaling strategies. The paper also surveys applications in manipulation, navigation, and autonomous driving, summarizes datasets, benchmarks, and metrics, and discusses key challenges such as action alignment, spatial consistency, long‑horizon memory, and efficient inference.