Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in
Embodied AI 101 artwork

Podcast

Embodied AI 101

Shaoqing Tan

Technology · 460 episodes
Follow

Guests who follow get this show's guest call alerts.

Subscribe+ Add a credit

Embodied AI 101 tracks the latest research at the intersection of AI and physical intelligence.Robotics, manipulation, world models, and the path from digital intelligence to embodied agents - each episode breaks down recent papers and advances in the field.We examine the methodology, real-world implications, and path to deployment. Covering robot learning, perception, control, and the fundamental challenge of grounding intelligence in the physical world.New episodes daily.

Listen on
  • Find on Spotify
EpisodesPeoplePostsSubscribe

Is this your show? Claim this page

Information

Creator
Shaoqing Tan
Category
Technology
Years active
2026
Episodes
460
Copyright
© 2026 Shaoqing Tan

Sign in to suggest a network

PodCred

The people behind this show

People get invited back because someone remembered they were great.

PodCred is the verified record of who made this show. A credit here means people who were there confirmed the work.

Hosts

  • S
    Shaoqing TanHostFeed-stated

Who is on this show?

Think of the guest you still think about. The producer who makes the host sound smart. The editor who saves every episode. If a name is missing here, that person is invisible everywhere.

Add a credit

Episodes

460 episodes
460 episodes
  1. Can One Human Video Teach Vision-Based Dexterous Manipulation?

    Nov 20, 202613 minAudio

    Can a single human video provide enough guidance for a dexterous robot to learn a useful manipulation skill? This episode explains the proposed division of labor between motion priors, residual reinforcement learning,...

    Credit the people on this episode

  2. ROAM: Can Vision-Language Risk Maps Improve Legged Navigation?

    Nov 20, 202613 minAudio

    How can a legged robot use a vision-language model to navigate trails without asking it to place every foot? This episode explains how ROAM connects semantic scene interpretation to reinforcement-learned navigation and...

    Credit the people on this episode

  3. Doppel: Can Rendered Poses Improve Cross-Embodiment Robot World Models?

    Nov 19, 202613 minAudio

    Can robot world models predict interactions better if we show them how the robot will move? This episode examines Doppel's rendered-pose conditioning, its experimental evidence, and what cross-embodiment prediction...

    Credit the people on this episode

  4. Elastic-WAM: Can Selective Imagination Speed Up Robot Manipulation?

    Nov 19, 202613 minAudio

    Can a robot spend less time imagining the future and still manipulate objects more successfully? This episode explains how Elastic-WAM turns future refinement into a learned computation-allocation problem. We examine...

    Credit the people on this episode

  5. ScaleVLA: Can a Frozen 552B VLM Improve Robot Manipulation?

    Nov 18, 202613 minAudio

    Can a frozen 552-billion-parameter vision-language model make robot manipulation more data-efficient? This episode explains what ScaleVLA connects, what its experiments measure, and what its results establish about...

    Credit the people on this episode

  6. VLA-0: Can Integer Action Decoding Simplify Robot Manipulation?

    Nov 18, 202613 minAudio

    Can ordinary numerical text turn a vision-language model into a capable robot policy? This episode explains how VLA-0 approaches robot manipulation through integer action decoding while retaining the base VLM's...

    Credit the people on this episode

  7. SimEX: Can Coding Agents Build Reusable Sim-to-Real Robot Skills?

    Nov 17, 202612 minAudio

    Can a coding agent learn manipulation skills in simulation and carry them onto a real robot with only a few physical trials? This episode explains SimEX's simulation-integrated auto-research loop and examines what its...

    Credit the people on this episode

  8. HT-Policies: Can Robust Regression Match Generative Robot Control?

    Nov 17, 202613 minAudio

    Can a single-pass regression policy match diffusion and flow matching in robot manipulation? This episode of *Embodied AI 101* explains how HT-Policies changes the treatment of action-prediction errors and why that...

    Credit the people on this episode

  9. CloudRobo-WM1.2: Can DualAct Make Robot World Models Practical?

    Nov 16, 202614 minAudio

    Can a robot world model generate useful actions without rendering the future every time? This episode explains how Huawei Cloud's CloudRobo-WM1.2 separates predictive representation learning, action generation, and...

    Credit the people on this episode

  10. RoboScript: Can In-Context Learning Beat Robot Manipulation Policies?

    Nov 16, 202613 minAudio

    Can a robot learn a useful manipulation procedure from one demonstration without task-specific fine-tuning? This episode explains how RoboScript implements World In-Context Learning by turning simulator practice into a...

    Credit the people on this episode

  11. GR00T 2: Can World-Action Models Make Manipulation Skills Portable?

    Nov 15, 202613 minAudio

    Can a robot learn dexterous manipulation from human experience and carry those skills into a different body? This episode explains GR00T 2's world-action architecture, cross-embodiment training, and the evidence behind...

    Credit the people on this episode

  12. Flash ALOHA: Fast Manipulation From 50 Demonstrations

    Nov 15, 202613 minAudio

    Can a robot learn fast, precise manipulation from just 50 demonstrations per task? This episode explains how Flash ALOHA connects direct human guidance, visuomotor imitation learning, and action-chunk execution to...

    Credit the people on this episode

  13. SSF: Can Bayesian 3D Belief Maps Improve Robot Object Search?

    Nov 14, 202613 minAudio

    How should a robot decide whether to look closer, explore somewhere else, or open a drawer? This episode explains how Spatial-Semantic Fusion connects those choices through a Bayesian 3D voxel belief map. We examine...

    Credit the people on this episode

  14. GSP-3D: Can Gaussian World Models Improve Bimanual Manipulation?

    Nov 14, 202613 minAudio

    Can predicting how a three-dimensional scene changes help a robot coordinate two arms? This episode examines GSP-3D's combination of Gaussian scene representations, diffusion-based action generation, and predictive...

    Credit the people on this episode

  15. Praxis-1: Can Video Pretraining Power Real-Robot Manipulation?

    Nov 13, 202613 minAudio

    Can a model pretrained to generate video become a useful real-robot policy? This episode explains Praxis-1's architecture, measured manipulation performance, and significance for robot foundation models. We cover three...

    Credit the people on this episode

  16. Gemini Robotics 2.5: Does Better Dexterity Mean Safer Robot Control?

    Nov 13, 202613 minAudio

    What does it take for a dexterous VLA to work safely beside people? This episode clarifies how Gemini Robotics 2.5 connects task reasoning, physical control, and robot learning - and what its safety evidence actually...

    Credit the people on this episode

  17. RoboCurate: Does Video-Action Verification Improve VLA Training?

    Nov 12, 202612 minAudio

    Can a realistic generated robot video still be bad training data? This episode explains how RoboCurate turns predicted robot actions back into visible motion, then uses that comparison to curate synthetic...

    Credit the people on this episode

  18. Can NTT's Physics-Aware World Model Prevent Manipulation Failures?

    Nov 12, 202613 minAudio

    This episode of Embodied AI 101 examines NTT's physics-aware world model for manipulation, separating the shelf-loading result from the proposed airport baggage application and explaining how an existing robot...

    Credit the people on this episode

  19. PEAOM: Better Sim-to-Real Table Tennis by Modeling Camera Errors

    Nov 11, 202614 minAudio

    Why does a table-tennis robot that succeeds in simulation miss the ball on hardware? This episode explains how PEAOM models the measurement process that turns a physical trajectory into a controller's estimated state....

    Credit the people on this episode

  20. Tactile VLA Avocado Sorting: Do Continuous Features Beat Labels?

    Nov 11, 202613 minAudio

    Should a vision-language-action robot receive tactile measurements as continuous features or as discrete labels? This episode explains the architectural choices behind that question and the evidence needed to interpret...

    Credit the people on this episode

  21. NEO-VLA: Why Visual Readout Depth Matters for Robot Manipulation

    Nov 10, 202613 minAudio

    Can an encoder-free vision-language-action model make robot foundation models more accessible? This episode explains NEO-VLA's architecture, its strongest experimental findings, and what its public release enables other...

    Credit the people on this episode

  22. SLA: Can Spatial Pretraining Improve VLA Robot Manipulation?

    Nov 10, 202614 minAudio

    Can a robot policy learn better three-dimensional control without running a geometric model every time it acts? This episode of *Embodied AI 101* examines how SLA uses spatial supervision during continual pretraining...

    Credit the people on this episode

  23. SLA: Does 3D Pretraining Improve pi0.5 Robot Manipulation?

    Nov 9, 202613 minAudio

    Can a robot learn better 3D manipulation without running an additional perception model at deployment? This episode examines how SLA extends π0.5 through spatially supervised continual pretraining and downstream...

    Credit the people on this episode

  24. Noetic: Can Collision-Aware World Models Make Robot Navigation Safer?

    Nov 9, 202613 minAudio

    Can a robot world model use imagined collisions to choose safer actions? This episode examines Noetic's reported safety-aware rollout loop and explains how it relates to latent planning, safe reinforcement learning, and...

    Credit the people on this episode

  25. Pi0.5 + SLAM: What 97% Success Means for Clothing Manipulation

    Nov 8, 202613 minAudio

    Can a vision-language-action model make clothing cleanup a useful autonomous household chore? This episode explains how Pi0.5 fits into a wheeled robot system combining SLAM, visual checking, collision detection, and...

    Credit the people on this episode

  26. Delta-0: Why Humanoid Foundation Models Need Correctable Control

    Nov 8, 202613 minAudio

    What makes a humanoid foundation model useful when its first attempt goes wrong? This episode examines how Delta-0 connects whole-body control, human-motion pretraining, and corrective learning - and what its...

    Credit the people on this episode

  27. PARMA: Can Pose-Aware Corrections Improve VLA Manipulation?

    Nov 7, 202614 minAudio

    Can a geometry-aware correction layer improve robot manipulation without retraining the VLA policy? This episode explains PARMA's proposed execution layer and the evidence needed to assess its pose-awareness claim. We...

    Credit the people on this episode

  28. LiMoDE: Can Dynamic Experts Improve Lifelong Robot Manipulation?

    Nov 7, 202614 minAudio

    Can a robot keep learning manipulation tasks while making better use of skills it already has? This episode explains LiMoDE through the relationship between computational specialization, skill retrieval, and memory. We...

    Credit the people on this episode

  29. A-0: Scaling Robot Manipulation with 16,000 Hours of Handheld Data

    Nov 6, 202614 minAudio

    Can handheld human demonstrations teach precise, long-horizon robot manipulation without collecting new robot trajectories? This episode explains AIRoA-0's learning pipeline, experimental evidence, and public release,...

    Credit the people on this episode

  30. VFlash: Can Parallel VLM Planning Deliver Real-Time Robot Control?

    Nov 6, 202613 minAudio

    Can parallel vision-language decoding make robot plans faster and more useful at the same time? This episode explains what NVIDIA's VFlash changes inside the planning pipeline and how to interpret its manipulation...

    Credit the people on this episode

  31. SADP: Can Subgoal-Aware Diffusion Improve Long-Horizon Manipulation?

    Nov 5, 202614 minAudio

    Can a diffusion policy learn both how to move and when to advance to the next manipulation stage? This episode examines SADP and the case for preserving subgoal structure in foundation-model-generated demonstrations. We...

    Credit the people on this episode

  32. VAMOS: Can One Navigation VLA Work for Both Wheels and Legs?

    Nov 5, 202613 minAudio

    Can one navigation VLA serve robots with radically different movement capabilities? This episode explains how VAMOS separates image-space route generation from robot-specific feasibility scoring, and what that division...

    Credit the people on this episode

  33. MJev-Spatial: Faster VLM Spatial Reasoning Without Generating Text

    Nov 4, 202614 minAudio

    Can a vision-language model make useful spatial judgments without generating an answer sentence? This episode of *Embodied AI 101* explains how MJev-Spatial turns that question into a concrete model design and...

    Credit the people on this episode

  34. VLM Latent Shaping: Can 3D Topology Improve Spatial Reasoning?

    Nov 4, 202613 minAudio

    Can a vision-language model improve spatial reasoning by changing the geometry of its hidden representations? This episode explains how Haoming Wang and Wei Gao turn object-level spatial structure into a fine-tuning...

    Credit the people on this episode

  35. ThinkWorld: Can Human Videos Teach VLA Manipulation Policies?

    Nov 3, 202613 minAudio

    Can action-free human video teach a robot more than visual recognition? This episode explains how NVIDIA and KAIST's ThinkWorld uses Chain-of-Dynamics reasoning and imagined future states to train vision-language-action...

    Credit the people on this episode

  36. Squint: Visual RL for Sim-to-Real Manipulation in 15 Minutes

    Nov 1, 202613 minAudio

    Can tiny camera images make sim-to-real reinforcement learning genuinely faster? This episode explains how Squint combines compact visual inputs with off-policy learning to accelerate robot-policy training. We cover...

    Credit the people on this episode

  37. BudSplat: Can Budgeted 3D Gaussian Splatting Preserve Scene Detail?

    Nov 1, 202613 minAudio

    Can a feed-forward 3D reconstruction model discard most of its Gaussians and still preserve scene detail? This episode explains how BudSplat makes primitive allocation part of reconstruction, from visual-token...

    Credit the people on this episode

  38. VoLo: Why VLA Tool Orchestration Helps Long-Horizon Manipulation

    Oct 31, 202613 minAudio

    Can a robot become more reliable by treating its vision-language-action policy as a tool rather than its entire controller? This episode explains how VoLo combines language-model orchestration, visual monitoring, and...

    Credit the people on this episode

  39. I3L: Why Corrections Beat More Demos in Mobile Manipulation

    Oct 31, 202613 minAudio

    Can a mobile manipulation policy become dependable by learning from the moments when a human takes over? This episode explains how I3L combines whole-body correction with iterative imitation learning, and what its...

    Credit the people on this episode

  40. World Labs Atlas and Marble 2: Do 3DGS Portals Advance Robotics?

    Oct 30, 202612 minAudio

    Vitre Experiment #6 presents a real-time portal into an AI-reimagined hallway, and this episode unpacks the distinct roles of World Labs' Atlas, the Marble 2 beta workflow, and 3D Gaussian Splatting, separating...

    Credit the people on this episode

  41. DR-WM-1.0: Can Physical Rehearsal Improve Outdoor Robot Navigation?

    Oct 29, 202614 minAudio

    Can a wheel-legged robot choose a safer outdoor route by rehearsing its actions before moving? This episode explains how DEEP Robotics' DR-WM-1.0 connects terrain estimation, physics simulation, and generated...

    Credit the people on this episode

  42. NeoWorld-3: Can Executable 3D Programs Improve Sim-to-Real Robotics?

    Oct 29, 202613 minAudio

    Can a robot turn pictures of its workspace into a world it can actually plan inside? This episode explains NeoWorld-3's approach to executable 3D programs and what its experiments reveal about their value for robotics....

    Credit the people on this episode

  43. Genie 4 World Simulation: Is It Ready for Robot Training?

    Oct 28, 202612 minAudio

    Google DeepMind's Genie 4 announcement brings native audio and Gemini-guided scene generation into the conversation about world models for robotics, and this episode separates the appeal of an editable, lifelike...

    Credit the people on this episode

  44. SuperMimic: Training Humanoid Motion Control on One GPU

    Oct 28, 202613 minAudio

    Can a differentiable world model make large-scale humanoid motion tracking a single-GPU experiment? This episode explains SuperMimic's training architecture, its comparison with SONIC, and what the experiments...

    Credit the people on this episode

  45. RynnUnify: Does Video Imagination Improve Robot Planning?

    Oct 27, 202613 minAudio

    Can imagined video help a robot plan and execute a multi-step task? This episode explains how RynnUnify connects scene understanding, video prediction, and robot control, and how to interpret the evidence behind that...

    Credit the people on this episode

  46. LaLa-World: Can Language Scale Latent World Models for Robots?

    Oct 27, 202614 minAudio

    Can natural language let one robot world model learn from incompatible control logs and ordinary video? This episode explains how LaLa-World unifies transition conditioning while keeping future prediction in visual...

    Credit the people on this episode

  47. SwarmACB: Why Behavior Selection Helps Swarm Sim-to-Real Transfer

    Oct 26, 202613 minAudio

    Why do reinforcement-learned swarms coordinate in simulation and struggle after deployment? This episode explains SwarmACB's proposal to learn behavior selection instead of direct wheel control. We examine three...

    Credit the people on this episode

  48. GVC: Why Action Modeling Helps Robot Reinforcement Learning

    Oct 25, 202613 minAudio

    Can the machinery that generates robot actions also teach a better critic? This episode of *Embodied AI 101* explains Generative Value Classifier, or GVC, and separates its value-learning contribution from its proposed...

    Credit the people on this episode

  49. World Labs Atlas: Is Single-Image 3D Generation Ready for Robotics?

    Oct 24, 202613 minAudio

    This episode of Embodied AI 101 examines World Labs Atlas through the practical questions behind single-image 3D generation: how a photograph becomes posed views and depth, what camera-controlled synthesis adds to a...

    Credit the people on this episode

  50. SpatialDex: Does 3D Video Pretraining Help Bimanual Manipulation?

    Oct 24, 202613 minAudio

    Can ordinary RGB human videos teach a robot to coordinate two dexterous hands? This episode explains how SpatialDex reconstructs spatial context from video and uses it to train bimanual manipulation policies. We cover...

    Credit the people on this episode

  51. Show-Harness: Can Small VLMs Replace Frontier Models in Robot Control?

    Oct 23, 202613 minAudio

    Can a small vision-language model take over robot control from a frontier agent? This episode explains what Show-Harness actually compresses, how its semantic action interface works, and where its strongest results come...

    Credit the people on this episode

  52. MolmoBot: Can Simulation Alone Deliver Zero-Shot Manipulation?

    Oct 22, 202613 minAudio

    Can a robot learn useful manipulation without real-world robot demonstrations? This episode of Embodied AI 101 explains what MolmoBot's simulation-first training achieves, how its data engine works, and what its...

    Credit the people on this episode

  53. VLA-Δ: Can World-Change Supervision Improve Robot Manipulation?

    Oct 22, 202613 minAudio

    Can a robot learn manipulation more reliably when training teaches it what should change in the world? This episode explains how VLA-Delta turns physical state transitions into supervision and examines what its...

    Credit the people on this episode

  54. Axis Robotics Browser Simulation: Does Sim-to-Real Manipulation Work?

    Oct 21, 202613 minAudio

    In this episode of Embodied AI 101, we investigate Axis Robotics' browser-based teleoperation platform and ask what it would take for distributed demonstration collection to improve manipulation policies on real...

    Credit the people on this episode

  55. HexaShell: Can Debugging Code Generalize RGB Robot Manipulation?

    Oct 19, 202612 minAudio

    Can a robot improve at manipulation by debugging code instead of updating neural-network weights? This episode explains how HexaShell turns robot experience into reusable tools, and what its RoboDojo score actually...

    Credit the people on this episode

  56. Niantic Spatial Gaussian Splats: Better Sim-to-Real Robot Training?

    Oct 19, 202612 minAudio

    Niantic Spatial's Places Library is the starting point for this episode of Embodied AI 101, which examines how Gaussian splatting, aligned collision meshes, and existing simulation tools fit together, why visual realism...

    Credit the people on this episode

  57. Gemini 4 Argon for Robotics: What the Benchmarks Actually Show

    Oct 19, 202612 minAudio

    This episode of Embodied AI 101 examines Google's Gemini 4 Argon robotics announcement, separating the demonstrated workflows from the broader promise by tracing how a calibrated camera view becomes a numerical motion...

    Credit the people on this episode

  58. GreenWAM: Can One World-Action Model Learn Across Robot Bodies?

    Oct 18, 202613 minAudio

    Can a robot foundation model learn a reusable connection between body motion and the scenes that motion creates? This episode examines GreenWAM's architecture, controlled experiments, and open benchmark release to...

    Credit the people on this episode

  59. 3DCodeVerse: Can Code Generation Build Better Worlds for Robotics?

    Oct 18, 202614 minAudio

    Can code-generating agents make 3D worlds easier to build, edit, and use in robotics? This episode explains 3DCodeVerse's training corpus, agent harness, and open-model experiments to show where its improvements come...

    Credit the people on this episode

  60. Motion- and Gravity-Aware 3D Reconstruction: Do Offline Priors Help?

    Oct 17, 202613 minAudio

    Can optical flow and gravity improve camera-based 3D reconstruction without becoming permanent parts of the inference pipeline? This episode explains the training design, benchmark evidence, and practical implications...

    Credit the people on this episode

  61. ASPIRE: Can Code Refinement Make Robot Manipulation Generalize?

    Oct 17, 202613 minAudio

    Can a robot improve by remembering how it fixed its own code? This episode explains how NVIDIA's ASPIRE turns robot execution feedback into reusable manipulation skills, and what its evaluations actually establish. We...

    Credit the people on this episode

  62. UAV-ESPrune: Can Visual Token Pruning Preserve UAV Reasoning?

    Oct 16, 202613 minAudio

    Can a UAV vision-language model discard most of its visual tokens while preserving the evidence needed for counting and planning? This episode explains UAV-ESPrune's selection strategy, how to interpret its results, and...

    Credit the people on this episode

  63. Gemini Robotics 2: How General Is Its Whole-Body Robot Control?

    Oct 15, 202630 minAudio

    Can Gemini Robotics 2 move beyond impressive manipulation demonstrations to become a dependable general-purpose robot? This episode separates Google DeepMind's demonstrated capabilities from the conclusions its public...

    Credit the people on this episode

  64. RoboJEPA to Long-WAM: Do New Embodied AI Models Generalize?

    Oct 15, 202614 minAudio

    This episode of Embodied AI 101 examines RoboJEPA, UNITAS, DreamTrue, VioLA, and Long-WAM as five different approaches to making learned representations useful for robot control, explaining how latent prediction, metric...

    Credit the people on this episode

  65. VLA-ACL: Can Token Pruning Speed Up Robot Manipulation?

    Oct 15, 202629 minAudio

    Can a VLA discard most visual tokens without changing what the robot does? This episode examines VLA-ACL's action-supervised pruning method and separates its demonstrated compression benefits from its deployment...

    Credit the people on this episode

  66. Juno: Do Predictive Latents Make VLA Manipulation More Robust?

    Oct 15, 202629 minAudio

    Can predictive world models improve VLA manipulation without interfering with action learning? This episode examines how Juno connects control-aligned JEPA representations, future-latent supervision, and deployment...

    Credit the people on this episode

  67. Can Rephrasing Fix VLA Language Sensitivity Without Retraining?

    Oct 15, 202628 minAudio

    Can changing an instruction make a frozen vision-language-action model more reliable without retraining? This episode examines what *Rephrase Before You Act* demonstrates, how its rewriting pipeline works, and where its...

    Credit the people on this episode

  68. Can CriticGPT Replace Hand-Designed Rewards for Robot Manipulation?

    Oct 14, 202629 minAudio

    Can a multimodal language model replace hand-designed rewards for robot manipulation? This episode examines what CriticGPT actually learns, how its video preferences become training rewards, and where the experimental...

    Credit the people on this episode

  69. Urbex: Can Agentic VLM Search Improve City-Scale 3D Grounding?

    Oct 14, 202629 minAudio

    Can an agentic vision-language model search a 3D city more effectively than a system that ranks candidate objects? This episode examines what Urbex improves, how its visual tools and reinforcement learning work, and...

    Credit the people on this episode

  70. VPP2: Does Better Video Prediction Improve Robot Manipulation?

    Oct 12, 202629 minAudio

    Can better video prediction make a robot manipulation policy more general? This episode examines what VPP2's experiments establish, how its training pipeline works, and where the causal story remains incomplete. We...

    Credit the people on this episode

  71. PointWAM: Does 3D World Modeling Improve Dexterous Manipulation?

    Oct 12, 202630 minAudio

    Can forecasting scene and hand motion in 3D make dexterous robot policies more reliable? This episode explains what PointWAM's geometric representation achieves, which experiments support its gains, and where the...

    Credit the people on this episode

  72. Mulligan: Can Failure-Focused Data Improve Robot Imitation Learning?

    Oct 11, 202628 minAudio

    Can a robot learn more from the same collection budget by revisiting the situations it gets wrong? This episode explains what Mulligan demonstrates about focused robot data collection - and why better learning per...

    Credit the people on this episode

  73. SOLE-R1: Can Video-Language Rewards Train Manipulation Policies?

    Oct 11, 202628 minAudio

    Can a robot learn manipulation from random actions using only rewards generated from video? This episode explains what SOLE-R1 demonstrates, how its reward model is trained, and where its generalization claims need...

    Credit the people on this episode

  74. Odyssey-3: Does a Physics-IQ World Model Transfer to Real Robots?

    Oct 10, 202630 minAudio

    Can a leading physics benchmark score translate into robots that need less task-specific training data? This episode separates Odyssey-3's documented video-generation results from its early evidence for embodied...

    Credit the people on this episode

  75. Long-WAM: Does Longer Visual Context Improve Real-Time Robot Control?

    Oct 10, 202629 minAudio

    Can longer video histories make robot manipulation more reliable without making control too slow? This episode examines Long-WAM to clarify when temporal context helps, what autoregressive pretraining contributes, and...

    Credit the people on this episode

  76. WR1.0: Does Whole-Body Humanoid Control Prove Generalization?

    Oct 9, 202629 minAudio

    Does WR1.0's clothes-folding demonstration establish a general-purpose whole-body humanoid foundation model? This episode separates the announced capability from the research evidence needed to judge autonomous...

    Credit the people on this episode

  77. WorldGuide: Better Video Planning, Not Proven Robot Control

    Oct 9, 202628 minAudio

    Can a goal-directed video world model genuinely help a robot plan, or does it mainly produce better procedural videos? This episode clarifies what WorldGuide demonstrates, how its planner-executor architecture works,...

    Credit the people on this episode

  78. R2SGen: Can Constraint Solving Improve Real-to-Sim Driving Scenarios?

    Oct 8, 202629 minAudio

    Can constraint solving turn a recorded traffic scene into useful simulation data without inheriting an LLM's geometric mistakes? This episode clarifies what R2SGen's guarantees mean, where its search improvements...

    Credit the people on this episode

  79. Humanoid Scooter Control: Does Sim-to-Real Reinforcement Learning Work?

    Oct 8, 202629 minAudio

    Can reinforcement learning trained in simulation let a humanoid control a real scooter? This episode examines what the reported transfer demonstrates, how the controller is structured, and why balance control should not...

    Credit the people on this episode

  80. ENPIRE: Can Coding Agents Improve Real-World Robot Manipulation?

    Oct 7, 202628 minAudio

    Can coding agents improve real-world robot manipulation without a researcher rewriting every experiment? This episode explains what ENPIRE automates, what remains engineered, and how to interpret its headline results....

    Credit the people on this episode

  81. GAS-Robo: Can Grid Actions Improve LLM-Driven Manipulation?

    Oct 7, 202629 minAudio

    Can a grid-action interface make language models better robotic manipulation planners, rather than just better selectors of predefined skills? This episode examines GAS-Robo's proposed architecture, its relationship to...

    Credit the people on this episode

  82. Four Keypoints, One Careful Pour: Connecting Language to Precise Robot Motion

    Oct 6, 202630 minAudio

    Bridging VLM and KMP: Enabling Fine-Grained Robotic Manipulation via Semantic Keypoints Representation

    Credit the people on this episode

  83. PEAfowl: Helping Two Arms Agree on What They See

    Oct 6, 202631 minAudio

    PEAfowl: Perception-Enhanced Multi-View Vision-Language-Action for Bimanual Manipulation

    Credit the people on this episode

  84. MotorMind: Giving a Frozen VLM the Controls

    Oct 5, 202629 minAudio

    MotorMind leverages a frozen VLM with mid-level prospective actions and asynchronous feedback to control robots zero-shot on manipulation tasks, requiring no task-specific training. This advances robot foundation models...

    Credit the people on this episode

  85. MAPF-World: Why a Little Foresight Can Beat a Bigger Policy

    Oct 5, 202629 minAudio

    MAPF-World: Action World Model for Multi-Agent Path Finding

    Credit the people on this episode

  86. Digit Leaves the Lab: The Real Lesson of 500 Demonstrations

    Oct 4, 202630 minAudio

    An end-to-end policy trained on ~500 teleoperation demonstrations combined with a sim-based whole-body controller achieves 80-90% success on unseen, chaotic real-world environments without any zero-shot fine-tuning....

    Credit the people on this episode

  87. VLS: Steering Frozen Robot Policies Without Retraining

    Oct 4, 202629 minAudio

    VLS steers frozen diffusion and flow-matching robot policies using vision-language models to handle out-of-distribution changes such as novel obstacles and lighting conditions, requiring no additional training data....

    Credit the people on this episode

  88. CAST: Making Language Matter to Robot Policies

    Oct 3, 202630 minAudio

    CAST: Counterfactual Labels Improve Instruction Following in Vision-Language-Action Models In this episode of Embodied AI 101, we explore "CAST: Making Language Matter to Robot Policies". We break down the research,...

    Credit the people on this episode

  89. When the Command Disappears: What Counts as Robot Intelligence?

    Oct 3, 202629 minAudio

    Robot Independent Intelligence: When Language Vanishes in Vision-Language-Action Models In this episode of Embodied AI 101, we explore "When the Command Disappears: What Counts as Robot Intelligence?". We break down the...

    Credit the people on this episode

  90. Fast Plans Are Not Enough: Does the Robot Actually Use Its Plan?

    Oct 2, 202631 minAudio

    Introduces a hierarchical VLA architecture that better bridges high-level language planning and low-level robot execution, reducing the gap between what a model plans and what it physically does. This advances robot...

    Credit the people on this episode

  91. Atlas's New Hands: Built to Learn, Not Just to Look Human

    Oct 2, 202611 minAudio

    Atlas is upgraded with 13 DoF dexterous hands featuring tactile sensing, direct drive actuation, and sim-to-real RL training, designed for real tool use and in-hand manipulation at scale. Represents a significant...

    Credit the people on this episode

  92. Astra Gets a Body: What Zero-Shot Humanoid Control Actually Transfers

    Oct 1, 202628 minAudio

    Astra demonstrates zero-shot open-ended pick-and-place on a physical humanoid robot, bridging simulation tooling directly to real-world execution without task-specific fine-tuning. This represents a meaningful step in...

    Credit the people on this episode

  93. Physical Coding: Robots That Check Their Work—and Keep the Fixes

    Sep 30, 202629 minAudio

    A new paper introduces self-evolving coding agents for robotics that represent task state and execution as code, enabling robots to verify, recover from failures, and evolve policies through real-world experience via...

    Credit the people on this episode

  94. SpatialClaw: Give the VLM a Geometry Notebook

    Sep 30, 202627 minAudio

    SpatialClaw is an NVIDIA spatial reasoning agent that uses Python-based code generation to perform 3D spatial reasoning, achieving +13.6 average improvement across 20 benchmarks with open vision-language models. The...

    Credit the people on this episode

  95. Real-Time EXPO-FT: Slow VLA, Fast Corrections, Better Control

    Sep 29, 202629 minAudio

    Async RL fine-tuning via EXPO-FT enables π0.5 to handle highly dynamic manipulation tasks such as ball balancing and striking in real time. The approach demonstrates that post-training with RL can unlock dynamic...

    Credit the people on this episode

Show moreShowing 95 of 460

Posts

No posts yet. Posts Embodied AI 101 shares on PodSubs, and what it publishes on Substack, YouTube and its own site, will show here.

Be the first to rate this show

Subscribe

Be on the front line of the show

Everything Embodied AI 101 sends its own people, first. Resources, announcements, behind the scenes, events, and giveaways, straight from the host. No account needed.

  • The resources the host shares with subscribers
  • Announcements before anyone else hears
  • Behind the scenes from the team
  • Events, invitations, and giveaways
  • Only what the host sends, nothing else
How often

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Report this show

Similar shows

  • Kim Komando Daily Tech Update artwork

    Kim Komando Daily Tech Update

    Also in Technology

    4,864 episodes

    Updated Nov 21, 2026

  • Rebuild artwork

    Rebuild

    Also in Technology

    561 episodes

    Updated Nov 21, 2026

  • Dad Reviews with @LaneVids (Svenska) artwork

    Dad Reviews with @LaneVids (Svenska)

    Also in Technology

    115 episodes

    Updated Nov 21, 2026

  • Tech Café : tech & IA artwork

    Tech Café : tech & IA

    Also in Technology

    561 episodes

    Updated Nov 21, 2026

  • Dad Reviews with @LaneVids (Deutsch) artwork

    Dad Reviews with @LaneVids (Deutsch)

    Also in Technology

    279 episodes

    Updated Nov 21, 2026

  • Dad Reviews with @LaneVids (Português) artwork

    Dad Reviews with @LaneVids (Português)

    Also in Technology

    278 episodes

    Updated Nov 21, 2026

  • Dad Reviews with @LaneVids (日本語) artwork

    Dad Reviews with @LaneVids (日本語)

    Also in Technology

    278 episodes

    Updated Nov 21, 2026

  • Dad Reviews with @LaneVids (Français) artwork

    Dad Reviews with @LaneVids (Français)

    Also in Technology

    279 episodes

    Updated Nov 21, 2026