SparseMax-dSTL proposes a smooth approximation of Signal Temporal Logic (STL) robustness based on sparse projection; its value always lies below the exact robustness and is therefore sound, and its error bound is independent of the number of samples and can be determined before optimization.
The authors propose the self-building laboratory (SBL) system, which can design, configure, and operate laboratory setups based on high-level scientific goals. Starting from an empty room or an existing self-driving laboratory, it fills in the laboratory layout, digital twin, robot skills, and experimental workflows. The system was demonstrated on a variety of robots and setups, and can scale from a single workstation to a laboratory with multiple robots working together. Its end-to-end capabilities were validated through a real-world quantum-dot synthesis experiment. The authors say this shifts the role of human researchers from manually configuring and programming device automation to specifying scientific goals.
Structural Affinity is an explicit regularization method for cage-based deformation (CBD). It uses a two-cage structure and an affinity matrix precomputed with mean-value coordinates to couple the motion of inner and outer control cages, so that control handle displacements propagate smoothly to neighboring handles, greatly reducing and often eliminating cage self-intersection and mesh distortion, while adding only linear overhead during deformation.
The paper releases TriSAR-3D, a trimodal person detection dataset collected with a 128-channel LiDAR, an RGB camera, and a thermal camera in indoor and outdoor settings at different times of day, covering diverse poses including lying down; it also proposes a fusion method, PaintFusion.
The paper proposes EmbodiedMidtrain, which bridges the gap between vision-language models (VLMs) and vision-language-action models (VLAs) through mid-training. The authors first characterize the data distribution gap between the two, finding that VLA data occupy compact regions largely separated from the VLM distribution. They then build a mid-training data engine that uses a lightweight learnable proximity estimator to select the candidates most aligned with VLA from a large VLM pool, producing the EmbodiedMidtrain-1M dataset for mid-training VLMs.
This study trains blind assembly policies that observe only joint positions, velocities, and a short history of their own actions, without using visual, force, torque, or tactile sensing, and uses reinforcement learning in simulation to complete peg insertion, gear meshing, and nut tightening. The peg insertion policy transfers zero-shot to the real world and successfully inserts pegs of different shapes and sizes in 73% of trials; controlled experiments show that the blind policy generalizes better to unseen part shapes and sizes but may be slower.
MOSS (Multimodal Origami Swamp Robot System) is a compact amphibious robot platform with 3D-printed origami wheels that switch between contracted and expanded configurations via a crank mechanism, simultaneously changing wheel width, wheel diameter, and fin orientation.
The paper proposes PRIME, a progress estimator for generating dense reward signals in reinforcement learning for robotic manipulation, learning precise task progress estimation from imprecise demonstration data. PRIME builds a weighted graph over observed states, connecting states across trajectories based on spatial proximity and visual consistency; edge weights encode spatial distance as transition cost, and shortest-path search approximates the geodesic distance to success, whose changes can distinguish progress, stagnation, and regression without assuming that progress is monotonic over time.
The arXiv paper Learning Robot Metareasoning proposes modeling the determination of the optimal computation budget as a semi-Markov decision process (SMDP), in which a single decision jointly determines the robot's action commitment duration and the time allocated to the solver. The authors trained a reinforcement learning policy conditioned on the current state, the executing plan, and low-cost planner statistics. Across three task types—ground robot navigation, agile quadrotor flight, and contact-rich manipulation—the policy adjusts its budget according to scene demands and outperforms fixed-budget baselines.
TransNav proposes a Transformer-based end-to-end navigation method for quadrotors that achieves obstacle avoidance in environments with mixed static and dynamic obstacles. The method uses a patch-based point-cloud encoding scheme, integrates perception and navigation into a single architecture, and directly predicts a finite-horizon action sequence and executes the first action. It also introduces a lightweight point-mass dynamics model that optimizes trajectory-level costs through numerical gradient propagation, enabling unsupervised training without expert demonstrations.
The paper integrates FOUND-IT into a mobile manipulation robot platform, validating that foundation-model-based 3D scene graphs can support object change detection, multiple mapping sessions, navigable path determination, and object grasping in scene rearrangement tasks. Based on rigorous real-robot experiments, the authors provide an analysis of the effectiveness and limitations of a monocular, foundation-model-driven 3D scene graph construction system. The code is open-sourced in the MIT-SPARK/FOUND-IT repository on GitHub.
The paper proposes Hindsight Inverse JEPA Planning (HIJP), which uses a goal-conditioned inverse model to predict actions directly from current and goal observations, replacing costly test-time action-sequence optimization in latent world models.
The TRACO framework enhances targeted visual attacks on vision-language-action (VLA) models through closed-loop optimization. Its Execution-Driven Supervision rebuilds supervision signals based on the states actually reached under attack, while Progress-Guided Selection concentrates updates where task progress stalls and maintains supervision of prior behavior. Experiments across multiple VLA models and simulated environments show that TRACO improves task redirection success rates, highlighting the security implications of closed-loop attack optimization.
This study proposes an embodiment-aware planning method in which two Franka FR3 arms equipped with different simple grippers (a suction cup and a parallel-jaw gripper) cooperate to complete manipulation tasks that a single gripper cannot perform reliably. Given a desired object motion specified by a human demonstration, the planner optimizes arm-trajectory assignments over a finite horizon, determining segment by segment whether single-arm manipulation or a coordinated dual-arm handover is required, while jointly considering heterogeneous grasp feasibility, motion constraints, manipulability, and cross-gripper compatibility; it also computes globally consistent assignments and adds explicit switching regularization.
This paper proposes a reason-guided preference-based reinforcement learning (PbRL) framework that learns a human reward model from trajectory-level feedback, introducing reason tags that explain why a behavior is preferred in addition to preference labels. Compared with preference-only reward learning, this method provides additional supervision for learning human-aligned rewards and improves the interpretability of the policy. The learned human reward is combined with the environment reward for policy optimization, enabling the agent to improve task performance while producing more interpretable behavior that better matches human preferences and aligns with human intent.
The paper proposes a reinforcement learning-based whole-body control framework for quadruped loco-manipulation on complex terrain. The framework combines a hybrid task representation with task-prioritized whole-body learning to coordinate locomotion and manipulation while preserving task-consistent end-effector behavior. The authors also developed a lightweight quadruped mobile manipulation platform for validation, and real-world experiments successfully deployed on complex terrains including slopes up to 30 degrees, stairs, and deformable terrain.
FLoTS-SLAM proposes deriving motion uncertainty from the residual between optical flow and camera-motion-induced flow, for monocular RGB SLAM in dynamic scenes. The system first estimates camera pose using high-confidence pixels, then reprojects dynamic pixels to jointly optimize pose and depth, reducing the impact of object motion on pose and depth updates. On DAVIS, the average IoU of dynamic masks improves by 35% over advanced baselines; on Bonn and TUM, average ATE is reduced by 59% and 19%, respectively, relative to the SLAM backbone used. The source code will be made public.
HexaShell is a lightweight framework that lets coding agents control a dual-arm robot via the command line. It achieved a 39.7% success rate and a 48.0 progress score on all 42 RoboDojo tasks, a new best result for RoboDojo.
The research team achieved Baoding ball rotation on the Shadow Dexterous Hand Lite using only proprioception and binary tactile feedback, enabling zero-shot sim-to-real transfer. In a single continuous trial on the physical hand, it completed 225 180° rotations (112.5 full rotations) at about 0.36 rotations per second, with no vision or object-state information throughout. The authors also compared the roles of touch and pure proprioception, identifying the hand regions where tactile information is most critical.
AWAM (Active World-Action Model) introduces a virtual active-perception paradigm that actively plans observation positions and, before execution, imagines what will be seen along the planned trajectory. It treats the predicted task-related virtual camera motion as a structured query into a geometry-conditioned world model, yielding a reusable spatial prior that is compressed into a fixed memory and fused with live observations through cross-attention for diffusion-based action prediction.
The paper proposes ActiveAlign, a data adaptation framework that addresses the mismatch between the initial observation state produced by active camera motion and the manipulation policy's training distribution (first-frame distribution sensitivity): on LIBERO, directly prepending camera repositioning to pi_0.5 reduces the success rate from 46.3% to 11.5%.
The paper proposes an Interventional Memory Evaluation Protocol (IMEP) for vision-language-action (VLA) systems, used to determine whether memory truly influences robot decision-making, covering four questions: the net benefit of memory, the conditions under which it takes effect, changes to task outcomes, and whether the effects are specific to task-relevant facts.
The authors propose LAW, a closed-loop VLM-WAM framework that unifies language reasoning, action generation, and world modeling through semantic subtask progress, achieving an average success rate of 94.21% across 50 RoboTwin tasks.
arXiv has published a survey that systematically reviews the memory mechanisms of long-horizon robot Vision-Language-Action (VLA) models, dividing them by memory scope and persistence into five representative paradigms and one emerging frontier. The survey compares representative VLA memory systems across memory characteristics, architectural complexity, and open-source maturity, finding a trade-off in which greater architectural complexity is generally accompanied by lower open-source maturity. It also discusses key challenges in building full-lifecycle memory systems, including efficient memory management and lifelong knowledge evolution.
SkyAgent proposes Revisable Target Memory (RTM) to address the problem in continuous aerial search where, after evidence becomes contradictory, old candidates still trigger approach or verification actions (which the authors call false candidate persistence, FCP). RTM associates evidence with persistent candidates, suppresses repeated counting of correlated views, and lets candidates be revised among three states: Tentative, Confirmed, and Rejected.
CoEvolve is a framework that represents team coordination as executable code over fixed robot motor skills. Through verifier-guided co-evolution, multimodal foundation models instantiate coordination strategies, physical verification evaluates each executed transition, and rejected behaviors trigger structured revisions starting from an accepted checkpoint.
LY-GWM-RoboCasa-Human300 combines a Human300-fine-tuned GR00T-N1.5 action generator, an action-conditioned graph dynamics model, and a binary state-reward scorer; at inference it samples 4 action chunks and executes the candidate with the highest predicted-state reward.
The NeurIPS 2026 paper NestedVLA proposes using a nested hypernetwork to generate task-adaptive action-expert parameters for a target task, combining them with a frozen base VLA. This replaces parameter aggregation with parameter generation, mitigating interference in multi-task training.
SimpleVLN is a vision-language navigation (VLN) framework that preserves the native architecture, achieving success rates of 76.7% and 80.1% on the R2R-CE and RxR-CE splits, respectively, significantly outperforming existing methods.
ORD-WM proposes a 3D world model that learns object-centric rigid-body dynamics for robotic manipulation tasks. It uses a vector-neuron variational autoencoder supervised by signed distance fields to encode each object point cloud into a geometry-aware latent representation composed of 3D vector features, and uses an action-conditioned dynamics model to predict per-object SE(3) motion increments. It then analytically applies the predicted rigid-body transformations to the current latent state rather than generating an unconstrained next latent state, thereby separating persistent geometry from pose changes.
SCOPE (Selective Certification Of Program Edits) uses failed and successful executions as complementary evidence to guide program revision: frontier-guided program search annotates a shared diagnostic graph with execution evidence to distinguish verified progress from unresolved checks, while selective behavioral certification uses stateful API replay to restore recorded responses and mutations and checks whether edits preserve selected historical successful behaviors.
The paper proposes BeJEPA, a goal-conditioned humanoid behavior world model learned through unsupervised reinforcement learning from human motion priors and robot interaction, using a joint-embedding predictive architecture (JEPA) to learn a shared representation of body configuration and world placement.
GRASP proposes a training-phase prior injection framework for bimanual robotic manipulation. A physics-grounded flow-matching loss uses differentiable forward kinematics (FK) to supervise tool center point (TCP) poses and inter-arm relative poses, and the framework introduces a global query group distilled from DINOv3 together with local interaction queries supervised by Grounding DINO; the teacher models are removed at inference. Simulation and real-world long-horizon bimanual task experiments show that the two priors are complementary under different deployment conditions: the full model performs better in in-distribution deployment, while the distilled query branch is more suitable under strong visual randomization.
InstAgent proposes a recursively self-improving (RSI) agent operating system that uses agentic behavior trees (BTs) to explicitly model capability assignment, inter-robot dependencies, and handoff checks for heterogeneous robots, and updates collaboration programs from execution feedback while preserving successful revisions for reuse in later tasks. Experiments on 20 real-world multi-robot tasks with 3 high-level models show consistent gains in task success rate, first-attempt completion rate, and execution efficiency.
The paper proposes a gradient descent bio-inspired magnetic navigation (BIMN) method based on local gradient estimation with a Rao-Blackwellized particle filter (RBPF) for navigation in GPS-denied environments. The method does not rely on complex, bulky, and costly magnetic maps; gradient descent naturally draws the system toward the destination, enabling magnetic navigation without magnetic maps as a simple and memory-efficient approach.
A critical arXiv review systematically examines the architectures, physical adaptation, and validation pathways of foundation models in autonomous space robotics, focusing on orbital manipulation and proximity operations. The article notes that event cameras can provide timely motion observations under strong illumination contrast, that spiking neural networks offer a possible route to sparse local computation, and organizes validation as an iterative "simulation-ground-space" closed loop, using a satellite capture case to connect observation, approach, contact, and recovery. The full text is 23 pages with 7 figures, accompanied by a source list and evidence document.
The paper proposes an architecture-agnostic probabilistic formal framework based on stochastic kernels for uniformly describing embodied foundation models. The framework has two parts: characterizing models with conditional output distributions and action-realization kernels to induce task-conditioned robot policies, and connecting pretraining, post-training, and policy deployment. The authors instantiate it on OpenVLA and DreamZero, covering both autoregressive action-token prediction and joint video-action generation paradigms.
PASS is a trustworthy precision approach simulation system for fixed-wing aircraft CAT III integrity algorithm development. Within a unified spatiotemporal geometric framework, it integrates configurable airport environments, prescribed trajectories, inertial simulation, calibrated stereo images, and continuous runway-relative ground truth, supporting navigation and integrity research under controllable sensor errors and environmental variations.
VLA-Fabric assigns a complete vision-language-action (VLA) model to each robotic arm, preserving each arm's local perception and action ownership, and achieves coordination through explicitly learned interaction across policy boundaries. The paper systematically studies where interaction should occur, how peer information should be aggregated, and how interaction should be learned, finding that shared context, peer-specific latent variables, and action-stage interaction play complementary coordination roles, and reproduces this principle on structurally different VLA architectures.
The paper proposes defining symbol usefulness from the perspective of the solving algorithm: a good symbol should expose the discrete decisions that simplify solving the original planning problem. The authors first introduce the concept of solver-sufficient symbolization for finite-horizon piecewise-affine systems and design a counterexample-guided symbol invention algorithm, then extend it to optimization objectives whose energy functions are general neural networks. Experiments on simulated and real-robot tasks show that the resulting symbols reduce solving cost while maintaining solution quality, and improve interpretability and controllability.