Xiaoyu Han of University College London and Dongkun Han of the Chinese University of Hong Kong proposed SCET-CBF, a speed-calibrated soft guidance law for one-step dynamic navigation. It uses a short lateral escape maneuver to provide an allowable approach speed, and lets a quadratic program weigh the two effects of braking when choosing acceleration and steering.
MOTIO is an agentic skill engineering system that turns ordinary human action videos, realistic 3D assets, motion references, and reusable robot skills into executable humanoid loco-manipulation programs, without task-specific teleoperation or robot demonstration collection.
TinyFlow is a 1.26M-parameter warp-free optical flow network that replaces flow-dependent feature resampling with fixed-window correlation over constant offsets, allowing the entire network to compile as a single static INT8 graph.
The paper proposes an active reachability learning framework for learning the stochastic finite-horizon recoverability envelope of a given fallback policy, so that fallback can be triggered before an unrecoverable fall occurs. The method reframes data exploration as a reach-avoid problem, in which an uncertainty-guided explorer discovers unseen states while avoiding confidently unrecoverable regions, and combines temporal-difference and Monte Carlo objectives to improve in-distribution value prediction and uncertainty estimation. The authors evaluate on stochastic benchmarks and a 29-DoF Unitree G1 humanoid robot, reporting clear improvements in monitoring coverage and accuracy, and zero-shot transfer to real hardware.
The authors propose Tiny Recursive Policy (TRP), a parameter-efficient behavior cloning policy that repeatedly updates high-level and low-level latent states using a shared computation module at each control step before decoding an action sequence, and propagates the state to subsequent predictions.
The authors propose Neuromorphic Decision Dynamics (NDD), a decentralized, event-driven controller for safe navigation in congested environments, with guarantees on safety, liveness, robustness, and scalability. NDD requires no communication, preassigned priorities, or models of neighbors' objectives; by coupling internal decision and attention states to physical states and the relative motion observed locally, it enables agents to self-organize into local conflict groups and form a passage order.
The authors propose ORBIT (Orbital Reconfigurable Base-actuated Integrated Theta-motion Hand), a modular multi-finger robotic hand platform whose core is a rotating knuckle mechanism: fingers travel around a cylindrical palm via a belt drive, enabling finger-layout reconfiguration within 1 second without disassembly.
This review surveys recent advances in aware, self-evolving robots, with the goal of enabling robots to give humans explanations for the reasons behind their actions, plan failures, or perceptual anomalies. The authors note that deep learning alone cannot close this gap, because the reasons behind the behavior of black-box models are not accessible; an underlying architecture based on deeper reasoning is needed. When explanations arise from cognitive processes and are shared with and validated by humans, the robot should be able to incorporate them into its knowledge base to improve subsequent behavior.
The study proposes a modular perception-prediction-planning pipeline that propagates uncertainty as structured, interpretable quantities across modules and guarantees a feasible solution in every cycle within convex safe sets. In Monte Carlo trials at a four-way unsignalized intersection, the pipeline achieved 0% collisions and 0% solver infeasibility, with 12.7 ms per cycle; the deterministic planner and chance-constrained baseline had collision rates of 36% and 10%, respectively. Ablation experiments confirmed the necessity of each uncertainty source, and a hardware deployment on a Clearpath Jackal UGV verified real-time feasibility.
The authors propose B-Adapt and BVSLAM to address image saturation and lost visual correspondences in visual SLAM when automatic exposure fails to adjust in time under rapid illumination changes. B-Adapt captures adaptive exposure brackets to cover the scene's dynamic range while maintaining photometric overlap, and BVSLAM uses the multi-exposure sequence directly as constraints in factor-graph state estimation without relying on HDR merging.
A Korean team proposed the RAIWAY wheeled bipedal robot and a contact-informed control method. The robot is equipped with 20-inch wheels and can roll across irregular and discontinuous terrain. The method distills multi-contact events into a latent representation in simulation, and at deployment infers it online from proprioception alone, requiring no contact sensors. Real-world tests covered steep slopes, gravel, studded surfaces, and rocky terrain, and demonstrated continuous ascent on outdoor stairs with step heights up to 15 cm, outperforming baselines without the contact-informed mechanism.
LANav targets the open-vocabulary object goal navigation (OVON) task, in which a robot must find a target in an unfamiliar indoor environment using only its current first-person image. Its approach is to have the robot stop repeatedly reviewing past frames and instead organize its memory to support navigation decisions.
RLT proposes freezing the large VLA model and training only a lightweight actor-critic control head: the VLA handles visual and language understanding and coarse actions, while the small network corrects actions online during critical stages to make up the last few millimeters of precision. The two are connected through a bridging structure.
The study proposes Quality-Controllable Policy (QCP), which enables inference-time control over individual quality attributes such as execution speed and motion smoothness, as well as their combinations. To extend the controllable quality range beyond the initial demonstrations, the authors designed an iterative hindsight relabeling loop: collect online rollouts under target quality labels, add successful trajectories to the dataset, then relabel the accumulated trajectories in hindsight according to their measured quality and fine-tune the policy.
A digital fruit fly reconstructed from micro-CT scans achieves hexapod walking in MuJoCo; the tripod gait self-organizes from phase-oscillator phase-locking, with reflex correction rules and adhesion measurements added. Its compound-eye vision outputs 721 ommatidia per eye, and all conclusions come from complete simulations with fixed seeds.
Cortical Labs' biological computer CL1, built from living cells, was trained in a simulated 2D environment and then used to control the movement of a real robot dog. A paper from the company that has not yet been peer-reviewed shows that the neural network improved over time when continuously fed the same data, and Ferrante Neri of the Education University of Hong Kong considers this preliminary evidence of learning from limited experience. Cortical Labs says the biological computer may consume less energy than digital approaches, and its cells come from blood stem cells donated with donor consent.
Researchers developed a single control policy that lets Unitree's G1 complete ladder climbing, crawling, obstacle traversal, stepping stones, and recovery after falls by deciding the placement of nine different body parts. The robot can use its hands, elbows, knees, feet, and even pelvis for support rather than relying only on walking, achieving a 92.5% success rate on a simulated obstacle course, but how this flexibility performs in unpredictable real-world environments remains a bigger test.
Long-WAM from teams at NVIDIA, MIT, the University of Hong Kong, and UCSD, and RoboJEPA from Meta FAIR, Mila, and other institutions were submitted to arXiv on the same day, October 7, measuring the returns from adding visual history and from expanding training compute, respectively.
The Tsinghua University Student AI4S Association held a forum on October 9. President Li Luming said bluntly that "a lot of research work is already meaningless," while Nobel laureate Omar Yaghi proposed "choosing science" and described an autonomous running laboratory in which AI proposes materials solutions and robotic systems synthesize and characterize them. Several professors stressed the building of foundational databases: Gao Xiangyu of Peking University Cancer Hospital advocated "one person, one file" with multi-point sampling throughout, Tsinghua's Xu Zhiping used physical transmission to connect simple and complex systems, and Sinopec's Sun Lili warned that data errors amplify step by step along the chain.
Zhang Changshui and Cui Sen's team at Tsinghua University propose RoboCoach, an embodied intelligence system that uses the world model CoachWorld to run closed-loop trials of long-horizon tasks in imagination, identify the first weak subtask that times out, and request targeted human teaching for it, updating only the corresponding skill expert. Under a fixed demonstration budget, the success rate rises from 13.3% to 75.0% on Franka and from 40.0% to 83.8% on AgileX, while the uniform collection baseline reaches only 30.0% and 47.5%, respectively. On four unseen new task routes, the average success rate is 35.0%, compared with 0% for the shared-policy baseline.
Why it matters: The paper presents a complete experiment on using a world model to locate weak subtasks and improve success rates through targeted teaching; its approach to data budget allocation may be a useful reference.
Researchers from UIUC, Cold Spring Harbor Laboratory, Brown University, and the Max Planck Institute propose DSReg (Dependency-Sparsity Regularization), which can be added directly to an already-trained LeJEPA so that each learned dimension corresponds to a real-world variable, without reconstruction of observations, a decoder, or labels.
The Samsung RX Lab-China team proposes ε4P (Imperfection for Precision), which admits imperfect data into different noise intervals of flow matching according to the type of data defect, thereby turning data that is harmful when mixed directly into training into a performance gain.
Researchers from Tsinghua University and the National University of Singapore propose ParallelWorld, which adds multi-step candidate future rehearsal to high-level planning for embodied agents, raising task success rate from 56% to 83%. The method generates and verifies multiple candidate action branches in a rehearsal loop; during execution it takes only the first action of the selected trajectory, then replans based on new observations. Success rates were higher than the direct baseline across eight LIBERO-Pro settings, but adaptive search took on average about 5.8 times as long as direct execution.
DUET proposes separating a humanoid robot's manipulation policy from its balance controller and combining the two commands only at the robot interface, so learning new manipulation skills no longer requires retraining the gait controller. The authors say this removes the hidden cost of retraining the balance controller every time the manipulation policy changes; the paper, code, and a full summary are in the first comment.
A NVIDIA team, together with Simon Fraser University, proposes GATOR, which uses generative reconstruction as a base and multimodal agents for local refinement to reconstruct complete 3D objects with scene poses from casually taken photos. The method is trained only on synthetic scenes and, without any real-world fine-tuning, leads baselines on geometry, appearance, and pose metrics in real tabletop and indoor scans; without refinement, it takes about 0.5 minutes to outperform GPT-6-Astra running for 20 minutes. The authors also note that reliance on synthetic data, binding to commercial APIs, and stability under extreme poses remain open issues for deployment.
Researchers have developed a bionic robot called BeetleBot that mimics a beetle's body and limb structure and gait control, allowing it to move stably across complex terrain such as rubble, sand, and narrow crevices. Experiments show it can climb over complex obstacles and quickly recover its movement state after external impacts or being flipped over. The robot is expected to be used in post-disaster search and rescue, exploration of hazardous areas, inspection of industrial facilities, and environmental monitoring.
The authors propose the CATS method, which combines a temporal penalty with linear ramp-up to suppress action oscillations in deep reinforcement learning policies. The paper proves that the temporal penalty can bound the expected action differences between current states that share the same successor state, thereby producing a spatial smoothing effect. In simulations and real-world environments, the method is verified to significantly reduce action oscillations without degrading task performance, with very low computational overhead. The work was submitted to ICRA 2027.
GenTrack3 proposes an online multi-object tracking (MOT) framework that integrates deterministic and stochastic principles, using stochastic particle sets to compute target-to-detection matching costs in order to maintain robust tracking under uncertainty. The paper also introduces a new track-to-detection matching method that improves scalability as the number of targets grows and supports group tracking; tracking inference uses tracklets containing identifiers, states, velocities, track penalties, and track ages. The authors state that a reference implementation and baseline trackers have been released on GitHub.
The authors propose TouchSafeBench, a physically grounded benchmark built in Habitat 3.0 for evaluating the collision grounding capability of vision-language models (VLMs). It contains 2,940 simulated indoor co-presence episodes covering two scenario types, social navigation and social rearrangement, and provides synchronized multi-view RGB-D observations, top-down trajectory maps, calibrated camera metadata, and simulator contact labels.
The authors propose TopoGS-SLAM, which decouples refinement from capacity using coverage-triggered Gaussian initialization and bounded-topology optimization, turning mapping into a bounded-dimensional problem. On standard indoor benchmarks, its localization and reconstruction accuracy is not inferior to densification-based pipelines, while memory scalability and throughput improve significantly. The authors therefore conclude that in bounded-area static indoor scenes with sufficient multi-view coverage, Gaussian SLAM does not require iterative topology mutation.
GenTrack2 proposes a visual multi-object tracking method that jointly uses stochastic particle filtering and deterministic association to maintain identifier consistency under conditions where the number of targets is unknown and changes over time and the dynamics are nonlinear. The particle filter uses particle swarm optimization (PSO) to guide particles toward the modes of the state distribution, with a fitness measure that combines motion consistency, appearance similarity, and social-interaction cues with neighboring targets; deterministic association strengthens identifier consistency through a cost matrix that incorporates spatial consistency between particles and current detections, detection confidence, and track penalties.
DepTR-MOT proposes a category-agnostic DETR-based detector that embeds instance-level depth awareness into the detector through instance-level soft depth label supervision and dense depth map distillation, and outputs instance depth at inference to improve multi-object tracking under occlusion and close-proximity interactions.
The authors propose LENS, which accelerates ACEsplat-style scene memory workloads on an AMD KR260 FPGA/ARM platform, comprising an INT16 scene coordinate regression accelerator and a fixed-point ordered 3DGS renderer.
The authors propose ContactBench 1.0, which uses 36 reference tasks and more than 40 object assets to break contact-rich manipulation down into eight essential skills; the assets can be reproduced on consumer-grade 3D printers and each comes with a matching simulation model. They evaluate recent vision-only and visual-tactile policy learning methods on six benchmark tasks, and the results show that vision-only policies struggle to perceive or respond to contact events, while visual-tactile policies do not bring noticeable improvement. The benchmark will be open-sourced, providing physical and simulated assets and controlled task variations for evaluating the generalization ability of robotic systems.
The paper proposes a series-parallel humanoid leg design that combines a knee parallel mechanism, an ankle parallel mechanism, and a nonlinear knee-ankle coupling, with all actuators placed near the hip. The authors derive coupled nonlinear transmission models and validate them in simulation and on hardware. The 6 kg leg achieves an effective reflected mass of 0.37 kg at the foot end; the coupling passively assists ankle dorsiflexion to lift the foot when the knee flexes, and assists plantarflexion for push-off when the knee extends.
ArtifactArena is an open-ended evaluation platform that challenges models with hardware-software co-design: building working robots in a simulated arena and pitting them against each other. The platform guides models to improve their robot artifacts through three channels: text descriptions, physics simulator feedback, and gameplay data, and compares frontier models using Elo rankings from head-to-head tournaments. The authors are releasing the framework and tournament infrastructure for ongoing community submissions, creating a non-saturating long-term testbed.
The research team proposes DASH, a da Vinci adapter for serial-link and humanoid robots that mounts da Vinci Classic and Xi instruments onto general-purpose robots without modifying the instruments themselves. The two adapter types are compatible with robotic platforms of various load capacities, engage the instruments' native latches, and wirelessly identify inserted instruments to load the corresponding kinematics and coupling matrices. Teleoperation experiments validated the approach and its role in lowering the barrier to surgical robotics research.
The authors propose RMMBench, a mobile manipulation benchmark that requires robots to understand language instructions and execute long-horizon tasks in continuous space, with a "navigation-manipulation" task suite covering 70 typical task scenarios. Results show that leading vision-language models (VLMs) still face notable challenges in spatial localization for mobile manipulation tasks, and their spatial perception in long-horizon interactions needs strengthening. The work has been accepted to CoRL 2026.
The study proposes Embodied Agent Arena, evaluating ten VLM agents across five dimensions—geometry, spatial reasoning, affordance, task planning, and manipulation—with 1,000 cases, including the 168-case GeoProbe benchmark for camera, motion, depth, and scale estimation.
The authors propose WorldEcho, which evaluates whether action-conditioned world models truly follow commanded actions over a broader action distribution through visual integrity and SE(3) end-effector trajectory alignment. The diagnosis shows that models trained on expert data perform well on demonstrated actions but produce action-inconsistent or visually invalid generations on non-expert trajectories.