Learning methods

Articles

• 

  From Guided Play to Autonomous Dexterous Practice

  DSP behavior-clones a robot play policy from minutes of human-guided activity, deploys it to collect hours of interaction data, and trains a world model for planning and self-play policy distillation. Its authors report physical-robot results on pushing and in-hand reorientation, including a 69.6% success-rate advantage over hindsight behavior cloning and comparable success to reinforcement learning with 490 times less interaction data.

  Published Oct 16, 2026

• 

  A Robot Memory That Saves What the Present Cannot Explain

  TTT-RM trains an online fast-weight state on the part of task history that cannot be reconstructed from a robot’s current observation and bounded memory bank. Its authors report simulation gains and a three-minute, eight-stage physical stowing demonstration, but the evidence provided is limited to their preprint abstract.

  Written Oct 13, 2026

• 

  How Small Errors Derail Robot Imitation Learning

  Errors made by an imitation policy can accumulate until the robot drifts beyond the distribution covered by its training demonstrations, a covariate-shift failure mode linked to poor generalization. One real-robot chopstick study addresses single-step errors with added support at critical grasping stages and error accumulation with synthetic corrective labels, while a later preprint evaluates a policy-calibrated noise method in simulated environments.

  Written Oct 11, 2026

• 

  One Learning Loop for Robot Orchestration and Action

  Robo-COP proposes a shared improvement cycle in which deployment executions provide demonstrations, recurring failures guide policy fine-tuning, and revised skills must pass verification before the new policy is adopted. Its authors report gains on ten simulated tasks and three physical tasks.

  Written Oct 11, 2026

• 

  How Diffusion Policies Can Preserve Distinct Action Modes

  Independent action-noise pairing can mix diffusion paths and produce averaged denoising responses that suppress distinct demonstrated behaviors. A label-free training add-on instead assigns noise to actions to preserve relatively separate routes without changing architecture or inference, although the reported results come from one preprint.

  Written Oct 11, 2026

• 

  Train Robots Where the Challenge Is Just Right

  Success Guided Sampling prioritizes simulated task configurations around the boundary of a robot policy’s current abilities. The abstract reports that the method solved locomotion and assembly tasks that prior methods failed to solve and that distilled RGB-based policies transferred zero-shot to assembly tasks on real hardware.

  Written Oct 11, 2026

• 

  Virtual Views Give Frozen Robot Policies a Better Angle on Precision

  SpatialHarness synchronizes a simulated scene with real robot execution and renders complementary perspectives for a frozen multimodal policy. Its authors report gains on four physical manipulation tasks, including plug insertion and Tower of Hanoi.

  Written Oct 9, 2026

• 

  When does human help teach a robot, and when is it part of a test?

  A reading guide to human input during data collection, online training and evaluation, illustrated with LeRobot and Bee.

  Written Oct 7, 2026

• 

  How does reinforcement learning teach a robot through trial and feedback?

  How task-specific rewards guide robot learning, illustrated by bounded physical walking and grasping studies.

  Written Sep 27, 2026