The paper introduces Rollback, a method that uses action‑conditioned world models to backtrack to critical decision points, branch into alternative actions, and compare outcomes to attribute success to individual decisions. By turning world models into counterfactual engines, Rollback provides fine‑grained, action‑level supervision without a separate value model and integrates distributed world‑model search with asynchronous real‑robot execution. Experiments on block extraction, ball tossing, and tabletop curling show that Rollback improves success rates by over 10% compared to the strongest baseline and surpasses human performance on two tasks within eight hours of training.
Action-conditioned world models offer a scalable alternative to costly real-world interaction for improving robot policies. However, existing world-model-based methods largely treat world models as fo...
arXiv:2607. 02431v1 Announce Type: cross Abstract: Reinforcement learning (RL) can overcome the demonstration-coverage limitation of imitation learning (IL) by allowing robots to improve through trial-and-error interaction beyond the states observed in demonstrations.
arXiv:2602. 13977v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) promises to unlock capabilities beyond imitation learning for Vision--Language--Action (VLA) models, but its requirement for massive real-world interaction prevents direct deployment on physical robots.
arXiv:2608.24885v1 Announce Type: cross Abstract: Action-conditioned world models are increasingly used as learned simulators for policy evaluation and improvement, yet their effectiveness rests on a...
arXiv:2607. 15065v1 Announce Type: cross Abstract: Predictive world models enable robots to plan by imagining the outcomes of their actions, but their value for control hinges on generating many rollouts quickly.
Action-conditioned world models are increasingly used as learned simulators for policy evaluation and improvement, yet their effectiveness rests on an unverified assumption: generated futures faithful...
arXiv:2610.40565v1 Announce Type: cross Abstract: Realistic, fast robotics simulations could enable large-scale training of embodied AI models that transfer to the real world. However, despite decade...
Dream to Remember (D2R) is a continual learning framework that steers a frozen generalist robot policy toward new tasks while preserving previously acquired skills. It uses a pretrained world model to rehearse past tasks through imagined rollouts, eliminating the need for replay buffers or new data. Experiments on CALVIN, LIBERO, and a real Franka robot show that D2R can learn multiple tasks sequentially—often entirely in imagination—without catastrophic forgetting, outperforming traditional fine‑tuning methods.
RoboCoach introduces a world-model-guided coaching framework that uses imagined failures to direct demonstration requests and expert updates for robot manipulation tasks. The system, called RIDI, runs reusable skill experts within a shared action-conditioned world model and records the first failing subtask to decide which demonstrations to acquire and which adapters to refine. Experiments on simulation suites and real robots show that with only 150 additional subtask demonstrations, success rates increase dramatically, and the coached experts transfer effectively to unseen task compositions.
The paper introduces a feedback world model that updates predictions online using the robot’s observed next state after each action, thereby closing the loop between prediction and observation. This lightweight feedback state corrects future predictions without extra training or parameter updates, and the authors provide convergence guarantees under bounded dynamics residuals. Experiments on LIBERO‑Plus, Robomimic, and real‑world tasks show significant reductions in prediction error (up to 76.4%) and improved policy performance, even when applied to pretrained foundation models like Fast‑WAM‑Joint.
The paper introduces a reinforcement learning post‑training scheme that trains robot world models on their own autoregressive rollouts, using a contrastive RL objective adapted from diffusion models. It also proposes a training protocol that compares multiple variable‑length futures, a multi‑view visual fidelity reward, and demonstrates state‑of‑the‑art rollout fidelity on the DROID dataset, outperforming baselines on LPIPS, SSIM, and human preference tests.
arXiv:2606. 27475v1 Announce Type: cross Abstract: Robots trained on real world data tend to be imprecise, slow, and brittle to perturbations.