Tuesday, November 10, 2026
10 papers added this day. New papers stay for 14 days.
Tactile sensing 4
Dexterous manipulation 3
Imitation learning and data collection 1
Humanoids and mobile manipulation 2
Learning Heterogeneous Tactile Representation for General Dexterous Manipulation
November 2026
LingBot-Touch is one tactile encoder pretrained on 5,000 hours of piezoresistive, magnetic and visuotactile fingertip data. Its frozen tokens raise the success of four policies on 26 real dual-arm dexterous tasks by 14 to 25 percentage points.
Bridging Task Planning and Execution for Language-Conditioned Bimanual Manipulation
November 2026
A zero-shot VLM planner for a bimanual humanoid that annotates each object side with accessibility and reachability, schedules compatible actions on both arms, and validates every stage. On the real CENTAURO robot it clears side-by-side clutter at 50-85% success where a planner without that information scores 0%.
PCTac: Mechanochromic Photonic-Crystal Tactile Sensor for Dexterous Manipulation
November 2026
A camera-based fingertip whose contact surface is a photonic-crystal film that changes color under load, so the tactile image is a hue map of local stress. Fed to a Diffusion Policy as a hue-difference image, it succeeds in 68 of 90 real-robot rollouts, against 57 for a GelSight Wedge and 29 for vision only.
GAP: Generalized Adaptation Prior for Dexterous Piano Playing
November 2026
Instead of transferring a piano-playing policy from simulation, GAP transfers a network that predicts how a correction to an executed trajectory changes keystroke accuracy. On a tendon-driven hand it raises F1 by 3.6%-23% on three pieces after about 7 minutes each.
Kimodo-Policy: Learning Kinematic Action Models for Humanoid Loco-Manipulation
November 2026
Kimodo-Policy turns a frozen pretrained text-to-motion generator into a closed-loop humanoid policy by adding visual ControlNet branches and a hand head. It reaches 72.14% on multi-task HumanoidArena and 60% mean success on eight real-world loco-manipulation tasks with 92M trainable parameters.
DexForesight: Grounding Dexterous Policies with Implicit Spatial Foresight
November 2026
Train a world model that predicts 3D hand-point motion and visual feature changes from depth and hand kinematics, then align a dexterous policy's latent queries to it so the policy runs on RGB alone. Success rises from 49.4% to 60.5% on 11 DexJoCo simulation tasks and reaches 58.3% on three real-robot tasks, against 50.0% for the best baseline.
BAT-WAM: Bridging Action and Touch in World-Action Models for Dexterous Hands
November 2026
BAT-WAM tokenizes whole-hand touch and joint state separately and links them through an RGB-conditioned Transformer, so each can be reconstructed from the other before feeding a world-action model. On seven real tasks with a tactile-gloved dexterous hand it averages 79.3% success versus 66.4% for the baseline.
Tactile Sensors Are an Effective Substitute for Wrist Cameras in Contact-Rich or Occluded Manipulation
November 2026
A controlled study on a real arm with ACT policies and two low-cost tactile pads on the fingers. Touch without a wrist camera beats vision with a wrist camera on bag retrieval (81.7% vs 50.0%), plug insertion (46.7% vs 33.3%) and whiteboard wiping (71.7% vs 46.7%).
GUARD: Anomaly-Informed Q-Steering for Generative Robot Policies in Dexterous Manipulation
November 2026
GUARD steers a frozen flow-matching policy at inference time with the gradient of a Q-function trained on a sparse task reward plus a dense anomaly score, so actions stay where the policy is well supported. It reaches 96% average success on four Adroit tasks (21 points above the best baseline) and lifts real Allegro screwdriver rotation from 20% to 60%.
UniLatent: Cross-Embodiment Transfer via Latent Observation Alignment
November 2026
UniLatent renders two robots at matched end-effector poses in simulation and trains a source image encoder to reproduce the features a frozen DINOv2 gives the target robot, so a policy trained on one arm's demonstrations runs on another. Zero-shot success in simulation averages 76.0% against 44.5% for pixel translation, and a real xArm-to-Franka transfer succeeds 27/30 with 10 target demonstrations per task.
Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.