We have now recorded over a hundred discussions with the best researchers in the world on embodied artificial intelligence and robotics, covering topics like scaling robot learning, humanoid whole body control, tactile sensors, sim-to-real, and so much more.

For our hundredth episode, we brought on our newest co-host, Ruijie He, to discuss what we’ve seen and learned in the first hundred episodes. Which were our most-watched episodes? Which institution shows up the most? Who was our most frequent guest? What did we learn?

Find out on Episode 100 of RoboPapers, with Michael Cho, Chris Paxton, Jiafei Duan, and Ruijie He!

Learn More

https://robopapers100.com/

Part of the secret is that LDA-1B jointly learns forward dynamics, action prediction, and visual forecasting, all in a structured DINO latent space which avoids the pitfalls of redundant pixel-level prediction which isn’t necessarily aligned robot action. This approach works on both dexterous hands and simple robot grippers; it also generalizes across objects, tasks, and scenes.

Jiangran Lyu joins us to explain. Learn more on Episode 98 of RoboPapers, with Chris Paxton and Michael Cho!

Abstract

Recent robot foundation models largely rely on large-scale behavior cloning, which imitates expert actions but discards transferable dynamics knowledge embedded in heterogeneous embodied data. While the Unified World Model (UWM) formulation has the potential to leverage such diverse data, existing instantiations struggle to scale to foundation-level due to coarse data usage and fragmented datasets. We introduce LDA-1B, a robot foundation model that scales through universal embodied data ingestion by jointly learning dynamics, policy, and visual forecasting, assigning distinct roles to data of varying quality. To support this regime at scale, we assemble and standardize EI-30k, an embodied interaction dataset comprising over 30k hours of human and robot trajectories in a unified format. Scalable dynamics learning over such heterogeneous data is enabled by prediction in a structured DINO latent space, which avoids redundant pixel-space appearance modeling. Complementing this representation, LDA-1B employs a multi-modal diffusion transformer to handle asynchronous vision and action streams, enabling stable training at the 1B-parameter scale. Experiments in simulation and the real world show LDA-1B outperforms prior methods (e.g., π0.5) by up to 21%, 48%, and 23% on contact-rich, dexterous, and long-horizon tasks, respectively. Notably, LDA-1B enables data-efficient fine-tuning, gaining 10\% by leveraging 30% low-quality trajectories typically harmful and discarded.

Learn More

ArXiV: https://arxiv.org/pdf/2602.12215

Project page: https://pku-epic.github.io/LDA/

Github: https://github.com/jiangranlv/LDA-1B

Checkpoints: https://huggingface.co/Wayer2/LDA-pretrain

Jason Ma, Tianyu Li, Anh-Quan Pham, and Chet Bhateja joined us to tell us more. They show that as the amount of data they use in pretraining increased, they saw predictable, statistically significant gains on accuracy metrics on held-out data (data not seen during training). They go on to talk about what they learned, and show how this can be applied to many different problems.

Watch Episode #99 of RoboPapers, with Michael Cho, Chris Paxton, and Jiafei Duan today to learn more!

Learn More

Read the blog post:

Luis Felipe Casas and Yu Xiang come on to tell us about UHAS: the Unified Hand Action Space, a sphere-based action representation which enables cross-ebmodiment learning of dexterous manipulation skills.

Watch Episode #97 of RoboPapers, with Chris Paxton and Jiafei Duan, today to learn more!

Abstract

Robot manipulation policies are typically tied to specific robotic hand embodiments, limiting the transfer of learned behaviors across platforms with different kinematic structures. In this work, we propose the Unified Hand Action Space (UHAS), a sphere-based unified action representation for cross-embodiment dexterous manipulation. UHAS represents robotic hand actions as geometric deformations of a canonical sphere and uses a Cascade Inverse Kinematics (CIK) algorithm to map the shared representation to embodiment-specific joint configurations. Using reinforcement learning, we train dexterous manipulation policies directly in the proposed action space for in-hand cube reorientation tasks. We evaluate our method in both simulation and real-world experiments across multiple robotic hands, including the Allegro Hand, LEAP Hand, Shadow Hand, and MANO Human Hand. Experimental results demonstrate effective dexterous manipulation, zero-shot transfer to unseen hands, rapid finetuning across embodiments, and successful real-world deployment. Our experiments show that the proposed UHAS representation enables stable dexterous control and cross-embodiment policy transfer across robotic hands.

Learn More

ArXiV: https://arxiv.org/abs/2607.03570

Watch Episode #96 of RoboPapers with Michael Cho and Jiafei Duan today!

Abstract

Human demonstrations offer rich environmental diversity and scale naturally, making them an appealing alternative to robot teleoperation. While this paradigm has advanced robot-arm manipulation, its potential for the more challenging, data-hungry problem of humanoid loco-manipulation remains largely unexplored. We present EgoHumanoid, the first framework to co-train a vision-language-action policy using abundant egocentric human demonstrations together with a limited amount of robot data, enabling humanoids to perform loco-manipulation across diverse real-world environments. To bridge the embodiment gap between humans and robots, including discrepancies in physical morphology and viewpoint, we introduce a systematic alignment pipeline spanning from hardware design to data processing. A portable system for scalable human data collection is developed, and we establish practical collection protocols to improve transferability. At the core of our human-to-humanoid alignment pipeline lies two key components. The view alignment reduces visual domain discrepancies caused by camera height and perspective variation. The action alignment maps human motions into a unified, kinematically feasible action space for humanoid control. Extensive real-world experiments demonstrate that incorporating robot-free egocentric data significantly outperforms robot-only baselines by 51\%, particularly in unseen environments. Our analysis further reveals which behaviors transfer effectively and the potential for scaling human data.

Learn More

Project page: https://opendrivelab.com/EgoHumanoid/

ArXiV: https://arxiv.org/abs/2602.10106

Github: https://github.com/OpenDriveLab/EgoHumanoid

Instead, why not initialize the search based on previous actions? This allows for incredibly fast policy inference and in many cases improved generalization, generating high-quality predictions with sub-ms latency. Jindou Jia and Jianfei Yang join us to explain.

Learn more on Episode 95 of RoboPapers, with Michael Cho and Chris Paxton!

Abstract

Diffusion-based policies have recently achieved remarkable success in robotics by formulating action prediction as a conditional denoising process. However, the standard practice of sampling from random Gaussian noise often requires multiple iterative steps to produce clean actions, leading to high inference latency that incurs a major bottleneck for real-time control. In this paper, we challenge the necessity of uninformed noise sampling and propose Action-to-Action flow matching (A2A), a novel policy paradigm that shifts from random sampling to initialization informed by the previous action. Unlike existing methods that treat proprioceptive action feedback as static conditions, A2A leverages historical proprioceptive sequences, embedding them into a high-dimensional latent space as the starting point for action generation. This design bypasses costly iterative denoising while effectively capturing the robot's physical dynamics and temporal continuity. Extensive experiments demonstrate that A2A exhibits high training efficiency, fast inference speed, and improved generalization. Notably, A2A enables high-quality action generation in as few as a single inference step (0.56 ms latency), and exhibits superior robustness to visual perturbations and enhanced generalization to unseen configurations. Lastly, we also extend A2A to video generation, demonstrating its broader versatility in temporal modeling.

Learn More

Project Page: https://jingliangli.com/A2A_Flow_Matching/

ArXiV: https://arxiv.org/abs/2602.07322

Github: https://github.com/JIAjindou/A2A_Flow_Matching

Kamil Dreczkowski, Pietro Vitiello and Edward Johns join us to talk about their extensive study on how to teach robots new skills efficiently, while also generalizing to novel object instances, using a combination of task decomposition and retrieval, work that was published in Science Robotics.

Watch episode #94 of RoboPapers now, with Michael Cho and Jiafei Duan, to learn more!

Abstract

Humans are remarkably efficient at learning tasks from demonstrations, but today's imitation learning methods for robot manipulation often require hundreds or thousands of demonstrations per task. We investigate two fundamental priors for improving learning efficiency: decomposing manipulation trajectories into sequential alignment and interaction phases, and retrieval-based generalisation. Through 3,450 real-world rollouts, we systematically study this decomposition. We compare different design choices for the alignment and interaction phases, and examine generalisation and scaling trends relative to today's dominant paradigm of behavioural cloning with a single-phase monolithic policy. In the few-demonstrations-per-task regime (<10 demonstrations), decomposition achieves an order of magnitude improvement in data efficiency over single-phase learning, with retrieval consistently outperforming behavioural cloning for both alignment and interaction. Building on these insights, we develop Multi-Task Trajectory Transfer (MT3), an imitation learning method based on decomposition and retrieval. MT3 learns everyday manipulation tasks from as little as a single demonstration each, whilst also generalising to novel object instances. This efficiency enables us to teach a robot 1,000 distinct everyday tasks in under 24 hours of human demonstrator time. Through 2,200 additional real-world rollouts, we reveal MT3's capabilities and limitations across different task families. Videos of our experiments can be found on at this https URL.

Learn More

Project Page: https://www.robot-learning.uk/learning-1000-tasks

ArXiV: https://arxiv.org/abs/2511.10110

Science Paper: https://www.science.org/doi/abs/10.1126/scirobotics.adv7594

Github: https://github.com/kamil-dreczkowski/learning_thousand_tasks

Enter CHORUS by Ria Doshi, Jeanette Bohg, and team. They train a single VLA to control multiple, diverse robots, given only local observations and identifying information. This allows robots to collaborate on tasks like using a tape measure, handing over a book, or lifting a laundry basket.

To learn more, watch Episode #93 of Robopapers now, with Michael Cho, Jiafei Duan, and Chris Paxton!

Abstract

Multi-robot collaboration allows robots to efficiently take on a wide range of tasks, from moving a couch through a doorway to assembling structures on a construction site. However, achieving such coordination in mobile multi-robot settings remains challenging: centralized methods conditioned on the combined observations of a team scale poorly with team size, and decentralized methods that train one policy per robot often require explicit alignment procedures or information sharing at inference time to overcome partial observability. Our key insight is that the visuomotor priors of pretrained vision-language-action (VLA) models should enable reactive, decentralized collaboration from each robot's local observations alone, without these inference-time assumptions. We propose CHORUS, a framework that adapts a single VLA backbone to control diverse, multi-robot teams. At inference time, each robot runs an independent copy of CHORUS, conditioned only on its own observations and a robot-identifying prompt. In real-world experiments including mobile tape measurement, library book handovers, and laundry basket lifting, CHORUS achieves a 64% point improvement over decentralized, from-scratch models, improves reactivity to teammate behavior by 40% points, and outperforms centralized baselines. Together, these results show that a shared VLA backbone is capable of achieving decentralized multi-robot collaboration, without per-robot policies or inter-robot communication at inference.

Learn More

Project Page: https://chorus-model.github.io/

ArXiV: https://arxiv.org/abs/2606.12352

Kevin Wu, Irmak Guzey, and Dandan Shan join us to talk about HUG: Human Unversal Grasping. They collect a dataset of 1M frames (27.8 hours) of egocentric human grasping data, and use it to train a new flow-matching algorithm to predict hand pose. They can then predict human hand poses and retarget them to robot hands, allowing for a massive improvement on a wide variety of zero-shot robot grasping tasks in everyday scenes.

To learn more, watch Episode 92 of RoboPapers with Michael Cho and Chris Paxton today!

Abstract

Humans can grasp objects effortlessly, whereas multi-fingered robots are far from this level of generality. We argue that the most natural source of robot grasping data is from humans, who pick up thousands of objects every day. We present HUG, a flow-matching model that generates diverse human grasps for any user-specified object in a single RGB-D image captured from a stereo camera. Using smart glasses, we first collect 1M-HUGs, an egocentric dataset of human grasps spanning 1M frames (27.8 hrs) and 6,707 object instances across 41 buildings. Next, to model the distribution of natural human grasps, our novel flow-matching model fuses RGB and depth observations to output a grasp parameterized by wrist translation, wrist rotation, and MANO hand pose. Predicted grasps can be retargeted to various robot hands, enabling zero-shot grasping in everyday scenes. To standardize evaluation, we build a new simulated benchmark, HUG-Bench, of 90 unseen objects from five geometric categories and various sizes, with metric-scale 3D meshes. We evaluate HUG in the real world on the 30-object test set of HUG-Bench across multiple stereo cameras, robot embodiments, and household environments. HUG outperforms the state-of-the-art grasping baselines by +23% and +34% on our challenging object set. Code, data, benchmark, checkpoints, and an interactive demo are released on our website: this https URL

Learn More

Project Page: https://grasping.io/

ArXiV: https://arxiv.org/abs/2606.17054

Github: https://github.com/KevinyWu/hug

Yinpei Dai and Yuejiang Liu made RoboMME as a solution: it’s a large benchmark which shows 16 different robot tasks, like counting objects or mastering timing. They show 14 different memory-augmented generalist policies across these different benchmarks. It’s an incredibly thorough and interesting result, aimed at driving forward this core robotic capability.

To learn more, watch Episode 91 of RoboPapers with Michael Cho and Chris Paxton!

Abstract

Memory is critical for long-horizon and history-dependent robotic manipulation. Such tasks often involve counting repeated actions or manipulating objects that become temporarily occluded. Recent vision-language-action (VLA) models have begun to incorporate memory mechanisms; however, their evaluations remain confined to narrow, non-standardized settings. This limits their systematic understanding, comparison, and progress measurement. To address these challenges, we introduce RoboMME: a large-scale standardized benchmark for evaluating and advancing VLA models in long-horizon, history-dependent scenarios. Our benchmark comprises 16 manipulation tasks constructed under a carefully designed taxonomy that evaluates temporal, spatial, object, and procedural memory. We further develop a suite of 14 memory-augmented VLA variants built on the π0.5 backbone to systematically explore different memory representations across multiple integration strategies. Experimental results show that the effectiveness of memory representations is highly task-dependent, with each design offering distinct advantages and limitations across different tasks.

In HANDOFF, Lizhi Yang proposes a 10-D learned whole body controller, which can be converted to whole body actions, and which can be used by a VLM-driven agentic planner to perform complex, multi-step manipulation actions in the real world.

We then discuss how it’s possible to deploy such controllers in the real world, how to make them safe around people via controlled barrier functions and safety functions. This enables humanoids which can move safely through dynamic, crowded environments in the real world.

Learn more in Episode 90 of RoboPapers, with Michael Cho and Chris Paxton.

Abstract

For a humanoid robot to be deployed in the real world, the choice of command space (i.e., the interface between task planning and whole-body control) is crucial. Existing whole-body controllers typically demand dense kinematic or spatial references that planners struggle to synthesize from task semantics. We instead propose a compact, explicit interface that is intuitive, general, modular, and expressive enough for diverse loco-manipulation skills. To this end, we introduce HANDOFF, a single humanoid whole-body controller that follows this interface and is distilled via multi-teacher KL distillation under a context-conditioned gating scheme into a mixture-of-experts student from three complementary specialists: whole-body motion tracking with safety-filtered data, locomotion, and fall-recovery. On the Unitree G1, HANDOFF matches state-of-the-art velocity tracking and offers one of the largest robust manipulation workspaces. We further demonstrate hardware feasibility through multiple natural-language-driven task roll-outs, powered by a VLM-driven agentic planner with no task-specific data or controller fine-tuning.

Learn More

Project page for HANDOFF: https://lzyang2000.github.io/HANDOFF/

Code: https://github.com/lzyang2000/HANDOFF

ArXiV: https://arxiv.org/abs/2606.06493

Safe-SAGE on ArXiV: https://arxiv.org/abs/2603.05497

SHIELD on ArXiV: https://arxiv.org/abs/2505.11494

Zhengtong Xu and Yeping Wang tell us about their new work Contact-Grounded Policy (CGP). CGP predicts future robot state and tactile feedback, and predicts this into actions for a compliant robot controller so that a four- or five-finger robot hand can perform complex tasks involving precise manipulation, delicate grasping, and tool use.

To learn more, watch Episode #89 of RoboPapers, with Chris Paxton and Jiafei Duan.

Abstract

Contact-rich dexterous manipulation with multi-finger hands remains an open challenge in robotics because task success depends on multi-point contacts that continuously evolve and are highly sensitive to object geometry, frictional transitions, and slip. Recently, tactile-informed manipulation policies have shown promise. However, most use tactile signals as additional observations rather than modeling contact state or how their action outputs interact with low-level controller dynamics. We present Contact-Grounded Policy (CGP), a visuotactile policy that grounds multi-point contacts by predicting coupled trajectories of actual robot state and tactile feedback, and using a learned contact-consistency mapping to convert these predictions into executable target robot states for a compliance controller. CGP consists of two components: (i) a conditional diffusion model that forecasts future robot state and tactile feedback in a compressed latent space, and (ii) a learned contact-consistency mapping that converts the predicted robot state-tactile pair into executable targets for a compliance controller, enabling it to realize the intended contacts. We evaluate CGP using a physical four-finger Allegro V5 hand with Digit360 fingertip tactile sensors, and a simulated five-finger Tesollo DG-5F hand with dense whole-hand tactile arrays. Across a range of dexterous tasks including in-hand manipulation, delicate grasping, and tool use, CGP outperforms visuomotor and visuotactile diffusion-policy baselines.

Learn More

Project page: https://contact-grounded-policy.github.io/

ArXiV: https://arxiv.org/abs/2603.05687

Enter DexSkin: a soft, deformable electronic skin which can be applied across different surfaces and used to cover robot hands or fingers. Suzannah Wistreich and Baiyu Shi talk to us about their work building DexSkin, showing how it’s useful for policy learning, including online reinforcement learning, and how it' can be calibrated and policies transferred across sensors. They also open sourced their code and methods for building the sensors.

To learn more, watch Episode #88 of RoboPapers now, hosted by Chris Paxton and Jiafei Duan!

Abstract

Human skin provides a rich tactile sensing stream, localizing intentional and unintentional contact events over a large and contoured region. Replicating these tactile sensing capabilities for dexterous robotic manipulation systems remains a longstanding challenge. In this work, we take a step towards this goal by introducing DexSkin. DexSkin is a soft, conformable capacitive electronic skin that enables sensitive, localized, and calibratable tactile sensing, and can be tailored to varying geometries. We demonstrate its efficacy for learning downstream robotic manipulation by sensorizing a pair of parallel jaw gripper fingers, providing tactile coverage across almost the entire finger surfaces. We empirically evaluate DexSkin's capabilities in learning challenging manipulation tasks that require sensing coverage across the entire surface of the fingers, such as reorienting objects in hand and wrapping elastic bands around boxes, in a learning-from-demonstration framework. We then show that, critically for data-driven approaches, DexSkin can be calibrated to enable model transfer across sensor instances, and demonstrate its applicability to online reinforcement learning on real robots. Our results highlight DexSkin's suitability and practicality for learning real-world, contact-rich manipulation. Please see our project webpage for videos and visualizations: this https URL.

Learn More

ArXiV: https://arxiv.org/abs/2509.18830

Project Page: https://dex-skin.github.io/

Github: https://github.com/sdwistreich/dexskin

Datasets: https://huggingface.co/datasets/swistreich/dexskin

MolmoAct2 provides a foundation for open research into robotics. It is associated with its own open dataset, an open-data action tokenizer, and a reasoning variant which predicts depth tokens. And people have actually been using it across the community, running experiments in their own labs or homes.

Haoquan Fang and Jiafei Duan tell us more. Watch Episode 87 of RoboPapers, with Michael Cho and Chris Paxton, now!

Abstract

Vision-Language-Action (VLA) models aim to provide a single generalist controller for robots, but today’s systems fall short for real-world deployment. Frontier models are closed; open-weight alternatives are tied to expensive hardware; reasoning-augmented policies pay prohibitive latency for their grounding; and fine-tuned success rates remain below the threshold for dependable use. We present MolmoAct2, a fully open action reasoning model built for practical deployment, advancing its predecessor, MolmoAct along five axes. (1) MolmoAct2 is built on top of our new Molmo2-ER, a VLM backbone specialized for spatial and embodied reasoning, trained on a 3.3M-sample corpus with a specialize-then-rehearse recipe. (2) We release three new robot datasets spanning low-to-medium cost platforms: MolmoAct2-BimanualYAM Dataset, 720 hours of teleoperated bimanual trajectories that constitute the largest open bimanual dataset to date; MolmoAct2-DROID Dataset, a quality-filtered Franka subset of DROID; and MolmoAct2-SO100/101 Dataset, a quality-filtered SO-100/101 subset. (3) We train and release MolmoAct2-FAST Tokenizer, an open-weight, open-data action tokenizer trained on millions of trajectories across five embodiments. (4) We design a new VLA architecture to graft the discrete-token VLM into the flow-matching continuous-action expert via per-layer key-value (KV) conditioning. (5) we propose MolmoAct2-Think, an adaptive-depth reasoning variant that re-predicts depth tokens only for scene regions that change between timesteps, retaining geometric grounding at a fraction of prior latency. In the most extensive empirical study of any open VLA to date, spanning 7 simulation and real-world benchmarks, MolmoAct2 outperforms strong baselines including π0.5, while Molmo2-ER surpasses GPT-5 and Gemini Robotics ER-1.5 across 13 embodied-reasoning benchmarks. We release model weights, training code, and complete training data.

Learn More

Project page: https://allenai.org/blog/molmoact2

Code: https://github.com/allenai/molmoact2

ArXiV: https://arxiv.org/pdf/2605.02881v1

And check out our episode on the original MolmoAct:

RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning.

Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Abstract

Despite the sustained scaling on model capacity and data acquisition, Vision-Language-Action (VLA) models remain brittle in contact-rich and dynamic manipulation tasks, where minor execution deviations can compound into failures. While reinforcement learning (RL) offers a principled path to robustness, on-policy RL in the physical world is constrained by safety risk, hardware cost, and environment reset. To bridge this gap, we present RISE, a scalable framework of robotic reinforcement learning via imagination. At its core is a Compositional World Model that (i) predicts multi-view future via a controllable dynamics model, and (ii) evaluates imagined outcomes with a progress value model, producing informative advantages for the policy improvement. Such compositional design allows state and value to be tailored by best-suited yet distinct architectures and objectives. These components are integrated into a closed-loop self-improving pipeline that continuously generates imaginary rollouts, estimates advantages, and updates the policy in imaginary space without costly physical interaction. Across three challenging real-world tasks, RISE yields significant improvement over prior art, with more than +35% absolute performance increase in dynamic brick sorting, +45% for backpack packing, and +35% for box closing, respectively.

Learn More

Project Page: https://opendrivelab.com/RISE/

ArXiV: https://arxiv.org/abs/2602.11075

Tutor Intelligence is a full-stack robotics company: they build robot arms, they sell robot arms, they write the software and they train neural networks. Josh Gruenstein, Jesse Michel, Shiraz Khan, and Joe McCalmon join us to tell us more about how they scale both teleop data and human interventions from their teleoperators in order to train the policies they need.

Watch Episode #85 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Learn More

Blog post: https://tutorintelligence.com/blog/building-a-100-robot-data-factory-toward-factory-ready-ai

So why not do both? Robometer combines both progress and preference supervision, resulting in a stable, scalable, and highly general reward learning approach. Anthony Liang, Yigit Korkmaz, and Jesse Zhang join us to tell us more.

Watch Episode #84 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Abstract

General-purpose robot reward models are typically trained to predict absolute task progress from expert demonstrations, providing only local, frame-level supervision. While effective for expert demonstrations, this paradigm scales poorly to large-scale robotics datasets where failed and suboptimal trajectories are abundant and assigning dense progress labels is ambiguous. We introduce Robometer, a scalable reward modeling framework that combines intra-trajectory progress supervision with inter-trajectory preference supervision. Robometer is trained with a dual objective: a frame-level progress loss that anchors reward magnitude on expert data, and a trajectory-comparison preference loss that imposes global ordering constraints across trajectories of the same task, enabling effective learning from both real and augmented failed trajectories. To support this formulation at scale, we curate RBM-1M, a reward-learning dataset comprising over one million trajectories spanning diverse robot embodiments and tasks, including substantial suboptimal and failure data. Across benchmarks and real-world evaluations, Robometer learns more generalizable reward functions than prior methods and improves robot learning performance across a diverse set of downstream applications. Code, model weights, and videos at this https URL.

Learn More

Project page: https://robometer.github.io/

ArXiV: https://arxiv.org/abs/2603.02115

Code on Github: https://github.com/robometer/robometer

Not PointWorld, though. PointWorld is a 3D world model trained from real and simulated data which can perform a wide variety of manipulation tasks on a real robot, including grasping or handling articulated objects, all without any additional fine tuning. Wenlong Huang joins us to tell us more about what makes this work and how it’s different from other world models.

Watch Episode #83 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

Abstract

Humans anticipate, from a glance and a contemplated action of their bodies, how the 3D world will respond, a capability that is equally vital for robotic manipulation. We introduce PointWorld, a large pre-trained 3D world model that unifies state and action in a shared 3D space as 3D point flows: given one or few RGB-D images and a sequence of low-level robot action commands, PointWorld forecasts per-pixel displacements in 3D that respond to the given actions. By representing actions as 3D point flows instead of embodiment-specific action spaces (e.g., joint positions), this formulation directly conditions on physical geometries of robots while seamlessly integrating learning across embodiments. To train our 3D world model, we curate a large-scale dataset spanning real and simulated robotic manipulation in open-world environments, enabled by recent advances in 3D vision and simulated environments, totaling about 2M trajectories and 500 hours across a single-arm Franka and a bimanual humanoid. Through rigorous, large-scale empirical studies of backbones, action representations, learning objectives, partial observability, data mixtures, domain transfers, and scaling, we distill design principles for large-scale 3D world modeling. With a real-time (0.1s) inference speed, PointWorld can be efficiently integrated in the model-predictive control (MPC) framework for manipulation. We demonstrate that a single pre-trained checkpoint enables a real-world Franka robot to perform rigid-body pushing, deformable and articulated object manipulation, and tool use, without requiring any demonstrations or post-training and all from a single image captured in-the-wild.

References

Project page: https://point-world.github.io/

ArXiV: https://arxiv.org/abs/2601.03782

SimTooReal aims to address this through, unsurprisingly, sim-to-real learning. Kushal Kedia and Tyler Lum talk about how it works: they procedurally generate tool-like objects, and then train with the universal objective of moving objects around to different locations. This creates a general-purpose model which can manipulate various tools to perform a variety of tasks in the real world.

Watch episode #82 of RoboPapers, hosted by Michael Cho and Jiafei Duan, now to learn more!

Abstract

The ability to manipulate tools significantly expands the set of tasks a robot can perform. Yet, tool manipulation represents a challenging class of dexterity, requiring grasping thin objects, in-hand object rotations, and forceful interactions. Since collecting teleoperation data for these behaviors is challenging, sim-to-real reinforcement learning (RL) is a promising alternative. However, prior approaches typically require substantial engineering effort to model objects and tune reward functions for each task. In this work, we propose SimToolReal, taking a step towards generalizing sim-to-real RL policies for tool manipulation. Instead of focusing on a single object and task, we procedurally generate a large variety of tool-like object primitives in simulation and train a single RL policy with the universal goal of manipulating each object to random goal poses. This approach enables SimToolReal to perform general dexterous tool manipulation at test-time without any object or task-specific training. We demonstrate that SimToolReal outperforms prior retargeting and fixed-grasp methods by 37% while matching the performance of specialist RL policies trained on specific target objects and tasks. Finally, we show that SimToolReal generalizes across a diverse set of everyday tools, achieving strong zero-shot performance over 120 real-world rollouts spanning 24 tasks, 12 object instances, and 6 tool categories.

Learn More

Project page: https://simtoolreal.github.io/

ArXiV: https://arxiv.org/abs/2602.16863

Elvis Nava joins us to talk about mimic-video and Mimic Robotics. Mimic-ivdeo is part of a new class of video-action models, capable of achieving complex, dexterous bimanual robotic manipulation with relatively little robot data.

One of the key findings from mimic-video is that pretraining on webscale video allows robots to learn physics priors; as a result, policies train faster, generalize better, and are capable of more impressive dexterity, versus training on static images or image-language pairs as per a VLM.

Watch Episode #81 of RoboPapers with Michael Cho and Chris Paxton to learn more!

Abstract

Prevailing Vision-Language-Action Models (VLAs) for robotic manipulation are built upon vision-language backbones pretrained on large-scale, but disconnected static web data. As a result, despite improved semantic generalization, the policy must implicitly infer complex physical dynamics and temporal dependencies solely from robot trajectories. This reliance creates an unsustainable data burden, necessitating continuous, large-scale expert data collection to compensate for the lack of innate physical understanding. We contend that while vision-language pretraining effectively captures semantic priors, it remains blind to physical causality. A more effective paradigm leverages video to jointly capture semantics and visual dynamics during pretraining, thereby isolating the remaining task of low-level control. To this end, we introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations. The decoder serves as an Inverse Dynamics Model (IDM), generating low-level robot actions from the latent representation of video-space action plans. Our extensive evaluation shows that our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.

Learn More

Project page: https://mimic-video.github.io/

ArXiV: https://arxiv.org/abs/2512.15692