Stanford HAI

FBG whiskers can enable underwater robots to perceive hydrodynamic trails with remarkable accuracy, mimicking the sensory capabilities of harbor seals.

Safety evaluations reveal that 6-21% of successful robotic manipulation rollouts still violate safety specifications, underscoring a critical gap in current methodologies.

Simulation-based pre-training can drastically improve the dexterity of robotic hands, outperforming traditional training methods with just a fraction of real-world data.

HyMeS enables robots to efficiently manage long-horizon interactions, achieving a 14.5-point increase in task success without requiring extensive demonstrations for every task configuration.

Speed Tuning achieves over 2.4x speed-up in robotic manipulation tasks without the need for extra data collection, revolutionizing policy execution efficiency.

SmartMage redefines 3D scene understanding by dynamically selecting modalities, achieving state-of-the-art results while minimizing irrelevant computations.

VLA models can achieve a 66% success rate in contact-rich tasks by addressing precision and force failure modes, a leap from the previous 41% benchmark.

Robots can now adapt to drastic physical changes in real-time, maintaining stable locomotion even with a locked leg or added weight.

Robots can now learn to manipulate dynamic environments effectively from just one static demonstration, drastically improving performance and efficiency.

End-effector traces boost cross-embodiment transfer in robot manipulation, enhancing real-world task performance by 28% when leveraging simulation data.

Zero-shot optimization of robot designs leads to a staggering 70% reduction in tracking error, showcasing the potential of motion-conditioned co-design.

Renormalization can redefine how we understand and address the sim-to-real gap in robotics by leveraging effective parameters that capture omitted dynamics.

Revealing robot motion in video models can transform how we predict and control robotic actions, achieving high fidelity with minimal training data.

Value-Aware MARO prevents performance collapse in multi-agent systems during communication failures, achieving over 20% improvement in returns under high-attrition scenarios.

A unified framework that transforms how we build and debug asynchronous robot programs, ensuring reproducibility and systematic debugging across environments.

Scaling visuomotor context to 8K timesteps enables robots to master complex tasks and adapt in real-time, outperforming previous models by a staggering margin.

Task-dependent action frames can enhance robotic manipulation performance, with MoF outperforming traditional single-frame policies in both simulation and real-world applications.

Frontier foundation models can directly control robots through visual interfaces, achieving remarkable success without any fine-tuning on robot-specific data.

Pix2Act transforms complex 3D manipulation into a simpler 2D prediction task, leading to significant performance gains and robustness against camera variations.

AMP achieves millimeter-level precision in 3D manipulation by transforming action learning into a pixel classification challenge, drastically improving inference speed and success rates.

Faithful reasoning in VLA models can boost policy responsiveness to rare scenarios by 1.6x compared to state-of-the-art approaches, revealing a critical gap in current alignment strategies.

Freeform Preference Learning allows robots to be trained on nuanced human preferences, leading to a dramatic 38% improvement in manipulation tasks.

Robots can now learn new tasks on-the-fly from just one demonstration, revolutionizing how we teach machines to manipulate their environments.

Adapting pretrained policies with just a modest multisensory dataset can enhance robot manipulation performance across diverse tasks without sacrificing prior knowledge.

Synthesizing 48,000 interaction trajectories without human input enables a humanoid robot to learn complex loco-manipulation tasks effectively.

Touch is not just an add-on; it fundamentally enhances object representation, leading to dramatic improvements in physical property estimation and manipulation tasks.

Policies trained in SimFoundry's automated environments achieve up to 40% higher success rates in real-world tasks by leveraging affordance-preserving scene variations.

Pretraining through play can revolutionize how robots learn dexterous assembly, achieving 60% success in tight insertions with minimal contact clearance.

Complex manipulation capabilities can be achieved by dynamically composing simple behaviors, leading to unprecedented precision and adaptability in real-world tasks.

Primitive steerability in VLAs allows for autonomous skill acquisition, enabling robots to learn new tasks without human demonstrations.

Cloak enables VLA models to seamlessly adapt to new robotic embodiments without any additional training data, revolutionizing the way we think about robotic adaptability.

This vine robot can autonomously navigate and manipulate in complex environments, overcoming traditional control limitations with a robust vision-based approach.

Advanced Vision-Language-Action models can be dramatically compressed by up to 50% without losing performance, reshaping our approach to robotic manipulation.

DREAM-Chunk transforms action chunking by leveraging latent world models to enhance robustness against stochastic dynamics without the need for policy retraining.

SC3-Eval achieves a remarkable 0.929 Pearson correlation in evaluating robot policies, revealing critical insights into their real-world performance.

Action-view augmentation can transform how robots adapt to unforeseen obstacles, boosting manipulation success rates significantly.

Flow Reversal Steering transforms vague human commands into precise robotic actions, achieving up to 95% higher success rates in real-world tasks with minimal training.

Zero-shot sim-to-real transfer for articulated tool manipulation is now achievable with just a few clicks, revolutionizing how robots interact with complex objects.

Action-conditioned predictions from a compact latent model reveal that diffusion methods can dramatically outperform traditional regression in scene forecasting for autonomous vehicles.

Naively scaling test-time compute is wasteful; strategically allocating it with DIRECT can enhance embodied agent performance while slashing latency by up to 65%.

A single VLA model enables decentralized multi-robot collaboration, achieving a 64% performance boost without the need for individual policies or communication.

High-quality dense rewards can elevate robotic manipulation success rates from 50% to near perfection, transforming how robots learn from their environments.

LadderMan enables humanoid robots to climb ladders and manipulate objects with unprecedented robustness and adaptability in real-world scenarios.

A triangular roller tip mount significantly reduces friction and improves the performance of growing vine robots, enabling reliable sensor integration for complex tasks.

Training VLA policies without human demos is now feasible, with LEGS achieving better performance than traditional methods at a fraction of the cost.

Forget hand-crafted physics models – NeuROK learns to generate realistic object deformations directly from data, opening the door to more general and scalable 4D simulations.

Humanoid states, not low-level actions, are the key to unlocking text-driven control, enabling a diffusion model to generate more natural and semantically aligned behaviors.

Achieve near-perfect robotic manipulation with just 20 minutes of robot experience by smartly finetuning vision-language-action models with reinforcement learning.

Get the performance boost of expensive sampling-based RL policies for a fraction of the compute by learning to prune action candidates early in the diffusion denoising process.

Generate navigable, 3D-consistent simulations of real-world locations with arbitrary weather and dynamic object configurations using only geo-registered video data.