Due Diligence Research

Deep-dive industry reports, hardware BOM teardowns, data matrices

Editor's Picks

Featured Research

Last 7 days · Technical and industry impact
Featured paper Impact 87/100

Ego2Robot: Scalable Robot Data Synthesis from Egocentric Human Data

Ego2Robot is a scalable pipeline that converts egocentric human manipulation videos into robot training data through action retargeting, robot-arm visual synthesis, and multi-level quality curation, producing 18,561 hours of data across 15 robot morphologies — the largest ego-to-robot dataset to date. Joint pretraining on Ego2Robot-synthesized and robot data consistently improves out-of-distribution generalization across visual appearance, scene layout, embodiment morphology, and task semantics, with benefits validated on real-robot deployment.

First large-scale ego-to-robot data synthesis pipeline: 18,561 hours across 15 morphologies with disentangled OOD perturbation evaluation and real-robot validation; directly attacks the VLA pretraining data bottleneck. Novelty 27/30, evidence 22/25, industry value 23/25.
Featured paper Impact 86/100

Why Does Action Chunking Improve Behavioral Cloning Performance in Robotic Control?

This paper rigorously investigates why action chunking improves behavioral cloning. Experiments in simulation and the real world refute the dominant hypotheses (temporal consistency, horizon reduction, representation learning), showing the real benefits come from non-Markovian expressivity, reduced compounding error, and a newly identified effect called implicit ensembling. The authors match action-chunking performance without chunking by deploying policies as ensembles of randomized delays, and propose an explicit-ensemble policy class that significantly outperforms action chunking in many domains.

Refutes the dominant explanations of action chunking with rigorous sim+real experiments, identifies implicit ensembling as the key mechanism, and shows how to match or beat chunking with simpler single-action policies. Novelty 28/30, evidence 22/25, industry value 21/25.
Featured paper Impact 85/100

PhyAI: Real-Time Physical AI at the Edge, Scalable Rollouts in the Cloud

PhyAI is a unified Physical AI inference engine with a single runtime for vision-language-action (VLA) and world-action models (WAMs) across onboard, edge, and cloud deployments. Model adapters encapsulate architecture-specific conditioning, solver, cache, and output logic while graph execution, kernels, and memory management are shared, yielding 1.40x–4.65x speedups over the official implementations of pi0, pi0.5, GR00T N1.7, and MiniCPM-Robot, plus a 2.08x latency reduction on Cosmos3-Nano-Policy-DROID. The paper also introduces the control-time Roofline to distinguish inference-bound from environment-bound control.

Unified Physical AI inference engine running VLA and world-action models on onboard/edge/cloud with model adapters; 1.40x-4.65x speedups over official pi0/pi0.5/GR00T N1.7/MiniCPM-Robot; Cosmos3 2.46->1.18s on 8xH20; control-time Roofline analysis; open-source. Novelty 23/30, evidence 22/25, industry 23/25.

Latest Papers

Bimanual Manipulation Within an 8 GB Budget: Zero-Copy Sensing and Quantized ACT on an Entry-Level Jetson

industrial-arm

This paper runs a complete bimanual SO-101 pick-and-place system — sensing, policy, and control — on an entry-level Jetson Orin Nano Super (8 GB), using a desktop GPU only for offline training. A GStreamer/NVMM zero-copy pipeline cuts worst-case latency from 117.31 ms to 101.52 ms and peak single-core CPU from 98.0% to 77.0%; TensorRT conversion brings ACT inference from 114.02 ms to 17.93 ms in FP16 (6.4x) and 12.65 ms in INT8 (9.0x) with task success preserved (19/20, 18/20, 19/20). Two undocumented ACT findings: INT8 calibration quantizes the ResNet18 backbone but accepts zero of 145 transformer layers, and quantization need is conditional on action-chunking configuration.

GORDON: Graph-based Object-centric Rewards for Decomposition of Long-Horizon Manipulation

software

GORDON learns dense rewards from action-free video demonstrations by representing each scene as a graph of detected objects and spatial relations, then embedding those graphs with a self-supervised graph neural network into a task-aligned latent space. The resulting reward reveals stage-wise object-state transitions in long-horizon tasks, enabling automatic subtask discovery and sequentially composed policies. On seven manipulation tasks across MAGICAL and ManiSkill3 it reaches 74.4% average success on long-horizon tasks, roughly +35 p.p. over the best learned baseline and +25 p.p. over an oracle.

StableMimic: Smooth Human-Like Recovery for Humanoid Motion Tracking - Learning Beyond the Tracking Distribution for Structured Post-Fall Behavior

humanoid

StableMimic is a unified humanoid motion-tracking policy trained beyond the nominal tracking distribution so it can recover from falls. Perturbed resets around human get-up references expose prone, supine, and off-balance states; dedicated tracking and recovery experts are blended by a proprioceptive gate, and a hidden successor-state objective shapes human-like recovery without exposing reference identity to the deployed policy. On the retargeted LAFAN1 dance subset it achieves the lowest errors on all four tracking metrics among five methods, recovers 100/100 in matched push-to-fall trials, and is validated on real Unitree G1 robots.

WCM: A World Critic Model for Vision-Language-Action Reinforcement Learning

software

WCM replaces scalar-return critics in VLA reinforcement-learning post-training with a lightweight LeJEPA-based World Critic that jointly predicts future latent states and estimates values, giving the critic explicit temporal structure. Across 149 tasks on four benchmarks and seven real-world manipulation tasks, WCM delivers state-of-the-art in-distribution and out-of-distribution performance on Pi0, Pi0.5, and OpenVLA-OFT.

BWM: A Low-Cost High-Fidelity World Simulator for Robot Learning

software

BWM is an open-source, low-cost, action-conditioned world model that predicts future observations for robot manipulation, serving both as a data engine that augments imitation-learning data with action-aligned rollouts and as a policy evaluator for closed-loop assessment, risk anticipation, and policy ranking. It ranks first overall in the WorldArena Challenge across Track 1 and two Track 2 applications, with checkpoints, code, and interfaces released.

CLIFT: Turning Gemini Robotics On-Device into Humanoid Specialists via Non-Invasive Closed-Loop Iterative Fine-Tuning

humanoid

CLIFT shows how closed-weight robot foundation models can be specialized for agile, contact-rich humanoid tasks through managed supervised fine-tuning APIs. By converting deployment-time reward feedback into API-compatible supervised data, CLIFT pushes Gemini Robotics On-Device to near-perfect success after two flywheel cycles — without ever accessing weights, gradients, or losses.

PAC-MAN: Perception-Aware CBF-RL for Whole-Body Safety in Humanoid Dodgeball

humanoid

PAC-MAN couples control-barrier-function safety with deployment-realistic onboard sensing for whole-body humanoid evasion. A policy trained with training-time CBF guidance and a head-mounted camera only achieves near-oracle dodge performance, and zero-shot deployment on the Unitree G1 succeeds on 95% of real-world throws.

FasTac: A Curved Multispectral Vision-Based Tactile Sensor for High-Speed High-Precision 3D Shape and Force Perception

components

FasTac is a curved vision-based tactile sensor combining multispectral photometric stereo, dynamic-convolution force estimation, and FPGA acceleration in a compact form. It reduces depth error to 0.0415mm MAE, estimates three-axis forces with ~2.4-2.7% NMAE, and cuts processing latency to 1.09ms on FPGA.

TacWAM: Anchor-Guided World Action Model with Mechanics-Aware Tactile Prediction

software

TacWAM is a world action model that predicts future tactile states — appearance, dense force fields, and deformation flow — to supervise contact-rich manipulation learning, reaching 75.0% average success across four real-world tasks and outperforming the strongest baseline by 37.5 percentage points.

TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM

software

TurboVLA reformulates the conventional V→L→A VLA paradigm into a direct V+L→A mapping, eliminating the LLM as the central bottleneck. With only 0.2B parameters, it achieves 97.7% average success on LIBERO at 32 Hz inference with just 0.9 GB VRAM on an RTX 4090, matching or outperforming much larger VLA policies.

HumanCLAW: Can Vision-Language Models Act Through a Body?

general

HumanCLAW introduces an evaluation framework that decouples action decision-making from motor execution for embodied AI, benchmarking 9 state-of-the-art VLMs across 1,218 egocentric find-navigate-interact episodes in 41 scenes. The best model achieves only 16.8% — revealing that current VLMs fundamentally lack embodied self-awareness.

SymmGrid: Super-Scaling On-Robot Learning with Parallelized Symmetries and Egocentric-Exocentric Visual Perception

software

SymmGrid uses parallelized symmetry transformations to dramatically accelerate on-robot reinforcement learning. Real-world manipulation tasks achieve 1.37-2.17× wall-clock training speedup, with convergence in as little as 10.9-16.6 minutes for peg-insertion and cable routing, and up to 2.59× improvement in normalized AUC.

$\pi\mathbf{R}^2$: Reactive Real-time Flow Policies

software

πR² makes large-backbone action-chunking flow policies reactive and real-time by splitting conditioning into a fast proprioception channel (fresh every tick) and an asynchronously updated slow vision-language channel, combined with a latency-adaptive flow schedule. Applied to GR00T-N1.7, it replans at ~25Hz on an A5000, improving success by up to 30% in real-world tasks.

Pictura: Perspective-View Self-Play at Scale for Driving

amr

Pictura is a GPU-accelerated multi-agent driving simulator providing egocentric perspective rendering at every step, enabling the first large-scale driving self-play policy trained directly from perspective images without privileged observations. The resulting Alberti policy approaches privileged vectorized performance and transfers zero-shot to Waymo layouts.

HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone

software

HiFi-UMI raises the fidelity of robot-free UMI data collection to 3 mm end-effector accuracy using head-mounted stereo-inertial SLAM, native relative pose estimation, and wide-angle stereo cameras. Policies post-trained solely on this data match in-domain teleoperation across three backbone families, and a 4000-hour pretraining corpus lowers action error on unseen tasks by 41%.

Data Pyramid for Embodied Manipulation

general

A systematic framework organizing the embodied data ecosystem into a five-tier pyramid: real-robot data, UMI-style data, egocentric and exocentric data, simulation data, and general vision-language data. Analyzes recent embodied foundation models through the lens of their data recipes, examining how different sources are selected, aligned, and mixed during pretraining.

FeelWorld: Visuo-Tactile World Model for Hierarchical Contact Prediction and Planning

general

A hierarchical visuo-tactile world model that jointly predicts future visual latents and three tactile states (contact state, 3D tactile latent encoding force info, slip state). Uses contact-gated asymmetric attention to maintain visual-only prediction before contact and enable joint visuo-tactile prediction during contact. Achieves LPIPS 61% lower than visual baseline after 80-step autoregressive rollout and 81.7% zero-shot planning success.

Learning Reusable Hybrid Motion Priors for Humanoid Locomotion from Motion Imitation

humanoid

A three-stage pipeline that turns motion-imitation skills into a reusable Hybrid Motion Prior (HMP) for humanoid locomotion. An expert policy learns to imitate retargeted human motion-capture clips, is distilled into a frozen RVQ codebook architecture, and task-level policies then solve locomotion tasks by selecting discrete codebook entries. The velocity-tracking policy is deployed on a real Unitree G1 robot.

ViTacWorld: Scaling Visuo-Tactile World Models for Contact-Rich Robot Manipulation

general

ViTacWorld is the first visuo-tactile world model framework for contact-rich manipulation. It leverages public real tactile datasets and simulated environments to pretrain visuo-tactile-action trajectories, exploiting the fact that tactile signals exhibit a smaller sim-to-real gap than visual observations. It serves both as a data augmentation engine and a policy evaluation tool, generating physically meaningful rollouts that improve downstream tactile policies.

One Hand Watches The Other: Dynamic Multi-Agent Cooperation for Sample-Efficient Bimanual Manipulation in Dynamic Environments

industrial-arm

DynaMAC resolves the causal limitation of multi-stream policies in dynamic settings by treating the opposite arm as a dynamic task parameter, enabling unified dynamic manipulation and bimanual coordination without an explicit leader-follower relationship. It outperforms leading baselines by 35+ percentage points while requiring 20× fewer samples, and generalizes zero-shot from static demonstrations to dynamic environments.

Addressing the Orchestration Gap in Generalist Robots via Physical Agency

general

Pigey introduces a closed-loop physical agent orchestrator that decomposes goals into subgoals, commands low-level VLA policies, tracks outcomes, and recovers from failures — all without additional data or fine-tuning. It achieves 4x SOTA on LIBERO-PRO (12.8%→53.3%) and lifts frozen policies from near-zero to 90% on real-world reasoning-limited tasks, demonstrating the orchestration gap between motor skills and agentic reasoning.

AXIS: A Growable Community-Driven Data Engine for Scalable Robot Manipulation

general

AXIS is a scalable community-driven data engine and benchmark for robot manipulation that enables browser-based teleoperation, automatically generates and validates new manipulation tasks, and transforms community-collected demonstrations into training-ready data. The AXIS dataset contains 207 diverse tasks and 50K+ trajectories. Continual pretraining on AXIS improves π0.5 by 5.8% overall, outperforms RoboCasa365 by 37.3%, and exhibits consistent scaling with increasing data volume.

RL-MACRO: A Cybernetic Closed-Loop Intelligence Framework for Multimodal Adaptive Robotic Craniotomy

surgical

RL-MACRO is a closed-loop cybernetic framework for autonomous robotic craniotomy that couples multimodal perception (force and sound), adaptive decision-making via offline Implicit Q-Learning (IQL), and robotic execution. A CNN-LSTM observer reconstructs the hidden temperature state from partial sensory feedback (R²=0.939, MAE=1.717°C). A dual-head Actor dynamically coordinates feed rate, spindle speed, and cutting depth. Experiments on bovine ribs and ex vivo goat skulls validate robust perception, adaptive recovery from force/temperature excursions, and smooth execution on irregular surfaces.

Robostral Navigate: Monocular RGB-Only Visual Navigation with an 8B VLM

amr

Robostral Navigate is an 8B vision-language model for visual navigation that consumes only monocular RGB images — the most ubiquitous sensor across robotic platforms — and predicts waypoints by pointing to the next target in the current camera view. Operating purely in image space makes it robust to changes in camera intrinsics and scene scale, enabling zero-recalibration deployment across wheeled, legged, and aerial robots. It sets a new state of the art on R2R-CE (77.4% success rate) and RxR-CE (75.1%), surpassing the best monocular methods by 10.5 points and even depth- or multi-camera systems by 5.3 points.

DEED: A Data-Efficient Post-Training and Experience-Driven Learning VLA Framework for Retail Humanoids

humanoid

DEED is a systems-level approach to bridging the lab-to-store gap for humanoid VLA robots, validated on a supermarket chip-restocking task with a Unitree G1-Edu. It demonstrates that careful data design and targeted post-training with a single GPU can transform a policy that fails under naive fine-tuning into a competent real-world system, suggesting the lab-to-store gap is primarily a systems integration challenge.

HOST: Robots Acquire Manipulation Skills in Seconds from a Single Human Video

general

HOST enables a robot to acquire novel manipulation skills in an average of 29 seconds from a single human video, while retaining previously mastered skills. It achieves 62% average success rate across tasks — exceeding zero-shot baselines by 45% and even surpassing models fine-tuned on 50 robot demonstrations per task, requiring 50x fewer demonstrations and acquiring each skill 507x faster.

Diffusion ReRoll: Revisable Denoising for Robotic Sequential Prediction

general

Diffusion ReRoll introduces structured re-noising for revisable denoising over prediction horizons, enabling earlier and later segments of a sequence to revise one another. It achieves relative gains of 21% over Diffusion Forcing in guided planning on OGBench, 56.5% relative improvement over Diffusion Policy on LIBERO-10, and best action-video consistency in unified video-action prediction.

Masked Visual Actions: A Pixel-Space Control Interface for Unified World Modeling

general

This paper introduces Masked Visual Actions, a pixel-space control interface that expresses robot actions as partially revealed trajectories of an arbitrary entity in a video. Finetuned with only 15 hours of masked examples, a single checkpoint achieves strong visual fidelity, forward dynamics prediction, inverse modeling, and model-based planning across diverse scenes and multiple embodiments.

Eversion-Growing Robots Enable Safe Navigation and Endoscopic Imaging Within the Spinal Subarachnoid Space

surgical

This paper presents a 2 mm diameter eversion-growing robotic platform for low-friction navigation within the human spinal subarachnoid space. Validated through computational modeling, phantom experiments, and intact human cadaver studies, the robot achieved 65.2% reduction in mean interaction force and 48.0% reduction in peak interaction force compared to push-based insertion, with 150 mm of controlled intrathecal extension and no observable dural or neural damage.

Koopman DCM: Unstable Eigenfunctions as Data-Driven Representations for Legged Balancing

humanoid

This paper generalizes Divergent Components of Motion as Koopman eigenfunctions, learning unstable eigenpairs from only 1 hour of real robot data. On a real biped, the learned Koopman DCMs improve reference walking pattern tracking and provide state-based viability constraints when combined with model predictive control.

Importance Sampling and PCA for Finding Failures in Commercial Autonomous Vehicles

amr

This paper applies Adaptive Stress Testing (AST) and Diffusion-based Failure Sampling (DiFS) to a commercial autonomous trucking stack, combined with PCA-based statistical analysis for failure mode classification. Both algorithms discover simulated collisions in merge and cut-in maneuvers that traditional Monte Carlo simulation cannot reach.

RynnBrain 1.1: Towards More Capable and Generalizable Embodied Foundation Model

software

RynnBrain 1.1 introduces an embodied foundation model family spanning 2B to 122B parameters with contact-point prediction and native 3D grounding. It outperforms all proprietary and open-source models on VSI-Bench, MMSI, and RefSpatial-Bench, and demonstrates cross-embodiment VLA deployment on Unitree G1, Astribot-S1, and Tianji-Wuji robots.

Manifold-Guided Motion Planning for Tight Assemblies

industrial-arm

CMG-RRT proposes critical-manifold guided sampling-based planning, achieving 100% success on all tight assembly benchmarks including the first automatic solution of the Elk disentanglement puzzle, with a probabilistic completeness proof.

Handroid: Bridging Dexterous Hand and Humanoid

humanoid

Handroid is a compact 27-DoF reconfigurable robot that functions as either a dexterous anthropomorphic hand or a 0.33m desktop humanoid, validated on manipulation, locomotion, and long-horizon reconfiguration tasks.

Orbis 2: A Hierarchical World Model for Driving

software

Orbis 2 introduces a hierarchical driving world model with a two-stage training paradigm combining diffusion forcing and teacher forcing, achieving SOTA results in long-horizon generation fidelity, steering responsiveness, and internal representation quality.

AC-VLA: Robust Out-of-Distribution Action Execution via Compositional Learning

software

AC-VLA is a plug-and-play compositional learning framework that addresses trajectory overfitting and perceptual shortcuts in VLA models, achieving ~28% absolute improvement on compositional OOD tasks while maintaining near-perfect in-distribution performance.

RoboTTT: Context Scaling for Robot Policies — A New Scaling Axis for Foundation Models

general

NVIDIA and Stanford researchers propose RoboTTT, scaling visuomotor context to 8K timesteps — three orders of magnitude beyond prior work — enabling one-shot in-context imitation from human video, on-the-fly policy improvement, and full completion of a five-minute, ten-stage assembly task with 87% improvement over single-step baselines.

Stigmergic Graph Memory: An Environment-Aware Approach for Many-to-Many Multi-Agent Pickup and Delivery

amr

Stigmergic Graph Memory (SGM) records recent execution signals on warehouse nodes and edges to rank feasible endpoints and route preferences, achieving 20.5-36.7% throughput gains across 15 map-load conditions without altering collision constraints or planner validity.

Scaling Behavior Foundation Model for Humanoid Robots

humanoid

A systematic scaling recipe for Behavior Foundation Models in humanoid control, achieving 10-82% MPKPE reduction through motion tracking paradigm, strategic data synergy, and the Humanoid Transformer architecture.

DriftWorld: Fast World Modeling through Drifting

software

DriftWorld replaces diffusion with drifting generative models for single-step world model inference at 30+ FPS — 17x faster than diffusion baselines — achieving state-of-the-art decision-making across multiple robotic manipulation benchmarks.

AeroAct: Action-Centered World-Action Models for Language-Conditioned Quadrotor Flight

drone

AeroAct is the first world-action model (WAM) demonstrated on a real-world quadrotor, adapting a pretrained video diffusion Transformer to predict trajectory-action chunks from egocentric vision, proprioception, and natural language for closed-loop aerial navigation.

BridgeFlow: Fast and Robust SE(2)-Equivariant Motion Planning with Flow Matching

general

BridgeFlow achieves fast SE(2)-equivariant motion planning via lightweight task-centric canonicalization paired with flow matching, delivering up to 15x inference speedup and 2x higher valid trajectory rates over diffusion baselines.

Reflex: Real-Time VLA Control through Streaming Inference

software

Reflex enables real-time streaming inference for flow-matching Vision-Language-Action (VLA) policies by exploiting the timestep-invariance property, achieving 2.58x speedup and 50Hz stable streaming on standard benchmarks.

Open-AoE: An Open Egocentric Manipulation Dataset and Toolchain for Embodied Learning

humanoid

Open-AoE is a large-scale egocentric manipulation dataset with roughly 2,000 hours of real-world video, structured annotations, and an end-to-end toolchain designed to lower the data barrier for embodied AI research.