#Dexterous Manipulation (79)

GEN-1.5 Runs a Continuous Cosmetics-Kit Assembly

@GeneralistAI · 18 views · 2026-10-15

Robot foundation modelsDexterous ManipulationLong-Horizon Autonomy

Systematically Exploring the Capabilities of GPT-6 Astra as Embodied Policies

Evaluates GPT-6 Astra as an embodied policy across six domains: hybrid control reaches 48% on RoboDojo and 92% on RxR navigation, while direct in-hand control and dense locomotion references stay unreliable; token use and inference latency constrain practical control.

Galbot Team, Chen, Xuchuan, Cheng, Xiaoqian · 60 views · Sep 29, 2026

Embodied AIRobot ManipulationDexterous ManipulationSep 29, 2026

0:18

DexWeave: Learning Dexterous Humanoid Loco-Manipulation from Human Demonstrations

@ErenChenAI · 216 views · 2026-10-09

Mobile ManipulationHumanoidMotion retargeting

Temporal Visuo-Tactile Learning for Dexterous Grasp Stability

Most robotic grasping research picks a pose from vision and closes a parallel gripper, while humans rely on fingertip touch and almost never drop things. The LASR Lab at TU Dresden reframes the question as "is this multi-fingered grasp stable before we lift?" and answers it with data. They mount four Digit 360 sensors on the fingertips of a 16-DoF Tilburg Hand on a 7-DoF xArm7 and automatically collect 10,000 grasp trials over 200 objects (105 rigid, 95 deformable, 1.9-244 g), recording external RGB-D, proprioception and four tactile streams (camera, audio, IMU, pressure) throughout each grasp. Stability labels come from SAM 3 measuring post-lift object height and agree with manual verification on 92.34% of trials. The predictor consumes a 3-second window anchored at grasp initiation: 16-frame RGB and tactile clips go through pretrained VideoMAEv2-Base, 100-step proprioception and 224-step audio/IMU/pressure go through Transformers, fingertip features are fused across modalities, time and fingers by weight-shared factorized attention, and everything is aggregated by attention fusion into a post-lift stability probability. Under 5-fold object-disjoint cross-validation, vision+proprioception+touch reaches 83.55% and removing touch costs 3.99 points (79.56%). Tactile spatial resolution matters monotonically (79.40% at 1x1 up to 83.55% at 112x112), and freezing touch to its final frame costs 2.07 points, so the useful signal is high-resolution and dynamic. Deployed as an online lift-or-regrasp gate (threshold 0.95, up to five regrasps) on 20 objects excluded from training, the visuo-tactile gate reaches 82.0% success among executed lifts, 10.5 points above a non-tactile gate. The ~1.2 TB dataset is public on OPARA.

Ken Nakahara, Aleksei Buvailik, Prokhor Kotov · 358 views · Oct 7, 2026

Dexterous ManipulationTactile SensingVisuo-TactileOct 7, 2026

0:20

Tesla Optimus gets way too comfortable in the kitchen: dexterous hands are the real test for humanoids

@murzik52 · 376 views · 2026-10-01

TeslaTesla OptimusHumanoid

1:16

Sharpa Wave Folds a Paper Airplane in Public

@clankrmedia · 556 views · 2026-09-30

Dexterous ManipulationHumanoidSharpa

2:25

DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation

@berkeley_ai · 571 views · 2026-09-29

Dexterous ManipulationTactile SensingWorld Models

RLT: Physical Intelligence masters the last millimeter with RL Tokens

Physical Intelligence (Sergey Levine's team, Mar 19, 2026) present RL Tokens (RLT): freeze a pretrained VLA, attach an encoder-decoder that compresses its internal embeddings into a bottleneck RL token, then run online RL with a ~1M-parameter actor-critic on the real robot, refining only the critical phase (editing VLA action chunks rather than generating from scratch, anchored by a BC regularizer plus reference-action dropout). Across four sub-millimeter tasks the critical phase speeds up by up to 3x, screw success goes 20% to 65%, and half of Ethernet insertion episodes beat every human teleoperation demo - with just 15 minutes of real robot data.

BLOG

VLAOnline RLRLPDPhysical IntelligenceDexterous Manipulation

0:12

Sharpa Robotics Demo Draws Crowds: The Robot Plays Better Than Me

@jonstephens85 · 770 views · 2026-09-27

HumanoidDexterous ManipulationSharpa

0:31

DexTacWAM: Making Touch Part of the Robot World Model

@robotsdigest · 539 views · 2026-09-22

Tactile SensingWorld ModelsBimanual Manipulation

DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation

Dexterous manipulation depends on contact dynamics that are often only partially observable from vision. Recent World-Action Models (WAMs) couple predictive video world modeling with action generation, but remain largely vision-centric and therefore cannot directly model these contact dynamics. We present DexTacWAM, a visuo-tactile WAM that encodes each fingertip independently, aggregates the resulting features through a finger- and pose-aware tactile compressor, and injects the tactile latent into a video diffusion world model for joint visuo-tactile world modeling. Across six contact-rich dexterous manipulation tasks on a 22-DoF bimanual platform, DexTacWAM achieves the highest score on every task, averaging 70.6 versus 38.0 for the strongest baseline. Ablations attribute the gain to modeling contact evolution as part of the predicted world state rather than tactile conditioning alone: removing tactile world modeling reduces the four-task mean from 74.7 to 26.6 while keeping the same tactile features and action expert. After four hours of tactile-encoder adaptation with a frozen pretrained vision VAE, our continual vision-to-touch learning extends the pretrained video model to touch using roughly 100 demonstrations per task without tactile midtraining, while retaining visual prediction quality within 0.5 dB of vision-only counterparts. The compressor retains 89.4% of pre-fusion contact recall while enabling 2.26x faster training and 1.29x faster inference. Together, these results show that pretrained video priors can be extended to distributed multi-finger contact dynamics in a data- and compute-efficient manner.

Haoran Yuan, Zekai Wang, Boning Shao · 730 views · Sep 21, 2026

Dexterous ManipulationWorld-action modelsVisuo-TactileSep 21, 2026

Towards Human-level Dexterous Teleoperation

Humans reorient, translate, and regrasp objects within a single hand by orchestrating continuous contact transitions, yet existing dexterous teleoperation either retargets kinematics while ignoring contact forces and object inertia, or relies on generative action priors that drift in closed loop. TeleDexter formulates dexterous teleoperation as hand-object co-tracking: the operator specifies synchronized subgoal sequences of fingertip positions and object pose, and a single-stage reinforcement-learning controller learns in simulated contact how to realize them. A hybrid reward couples sparse subgoal reaching with dense tracking, geometry-aware two-stage retargeting converts mocap human hand-object data into physically feasible reference motions, and random action masking plus domain randomization enable zero-shot sim-to-real transfer. On seven reorientation and long-horizon tool-use tasks with a Franka FR3 and SharpaWave or LeapHand hands, TeleDexter averages 75.2% success and 87.1% task progress where baselines nearly uniformly fail, and 50 demonstrations per task behavior-clone into autonomous policies reaching 46.7-73.3% success.

Li, Puhao, Chen, Zeyuan, Wu, Yingying · 644 views · Jul 13, 2026

Dexterous ManipulationTeleoperationSim-to-Real TransferJul 13, 2026