Embodied AI 101

Shaoqing Tan

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.

Play

Open in

110 episodes
Avg 26 min
English

Episodes 110

Newest Oldest Longest

All20 Unplayed20 In progress0 Played0

Counted on this page — what you have heard stays on this device, so it is not something the list can be paged by.

September 22 · 28 min
Teach the Robot Where to Act—Then Make It Fast
We present a framework for assistive robot manipulation that addresses two fundamental challenges: efficient adaptation of large-scale models for scene affordance understanding and effective learning of robot actions by grounding...
- Transcript
September 22 · 28 min
HEAR: The Sound Your Robot Missed Between Actions
Humans and animals use sound as a crucial cue for interacting with the physical world, as acoustic events can reveal contact, completion, hidden contents, or process state. Embodied agents should leverage sound for manipulation tasks. This paper presents
- Transcript
September 20 · 29 min
VS-Splat: Learning Where 3D Detail Belongs
A new feed-forward Gaussian Splatting method that selectively processes voxels for efficient 3D scene reconstruction, targeting real-time perception stacks for embodied agents. Advances the efficiency frontier of 3D scene representation without per-scene
- Transcript
September 20 · 13 min
Spatial Search Comes to Quest — Beyond Pretty Gaussian Splats
Adapts a training-free Spatial Search prototype to Meta Quest's spatial data pipeline, enabling natural-language search over Gaussian Splatting and mesh scenes in real time on-device. Demonstrates practical deployment of 3D spatial intelligence for consum
- Transcript
September 19 · 30 min
Occamy-1.0: Teaching a 35B Agent to Finish the Job
Introduces a 35B cost-efficient model positioned at the Pareto frontier for collaborative long-horizon agent tasks, achieving performance competitive with significantly larger systems. Advances the capability of embodied and agentic AI to handle extended
- Transcript
September 18 · 12 min
Helix 2.5: The House Changed. The Policy Didn't.
Figure's Helix 2.5 humanoid robot enters 30 unseen Bay Area rental homes with no additional training and performs useful whole-body tasks for 4+ hours, demonstrating strong zero-shot generalization across diverse real-world environments. This is a notable
- Transcript
September 16 · 12 min
The Hard Part Was the Stack: PI's Robot Goes to Work
Physical Intelligence's mobile π robot runs fully autonomous for hours in a real production environment stacking boxes at Dandelion Chocolate, exposing generalization and reliability gaps between lab demos and production utility. Represents a significant
- Transcript
September 16 · 13 min
ZDTaichu5.0-9B: Better Spatial Reasoning, an Unfinished Edge Story
ZDTaichu5.0-9B is a 9B edge-deployable multimodal model built on a Qwen3.5 backbone with C-RADIOv4 vision encoder that leads open sub-10B VLMs on spatial-reasoning benchmarks including ViewSpatial and MMSI-Bench, while supporting embodied AI and tool-use
- Transcript
September 15 · 28 min
Light-Loco-Parkour: Learning the Skill Is Only Half the Problem
Light-Loco-Parkour presents a multi-skill distillation approach for learning perceptive whole-body parkour and locomotion skills deployable on real robots, enabling versatile agile movement across diverse terrains. The method advances embodied locomotion
- Transcript
September 15 · 29 min
OM-1: Human Skills, Many Bodies
OM-1 is a robot foundation model trained exclusively on human manipulation data (no teleoperation or robot demonstrations) that achieves near human-level dexterity and zero-shot generalization across tabletop arms, industrial arms, and humanoids, includin
- Transcript
September 14 · 12 min
Show-Harness Gives VLMs a Robot Keyboard
Uses discrete semantic actions paired with embodiment-specific interpreters to turn any pretrained VLM into a robot controller without a dedicated policy network or additional pretraining. Demonstrates strong zero-shot generalization across diverse tasks
- Transcript
September 14 · 29 min
BridgeVLA++: Keep the Geometry, Give the Robot a Memory
Renders point clouds as three orthographic 2D views fed into a pretrained PaliGemma VLM, predicting heatmaps to recover 6-DoF actions with temporal/spatial memory modules. Achieves 95.4% success on 13 real Franka tasks with only 3 demos per task, outperfo
- Transcript
September 13 · 28 min
The Sim-to-Real Gap Inside the Gearbox
Reinforcement learning (RL) has become a powerful tool for quadrupedal locomotion, and a sim-to-real approach is widely adopted to avoid hardware damage during training. However, the sim-to-real gap remains a significant challenge.
- Transcript
September 13 · 28 min
A Stable Grasp Isn't Enough: Giving 6-DoF Grasping a Purpose
Robotic grasping in cluttered scenes is dominated by methods that optimise either stability or a predefined task category, leaving a gap between 'can this grasp hold the object' and 'can this grasp serve the task'. This paper bridges that gap using founda
- Transcript
September 13 · 29 min
HiDream-O1-Embodied: A World Model Must Earn Its Actions
HiDream-O1-Embodied is a unified native architecture that ingests text, images, and video and directly outputs physical actions, positioning itself as a step beyond passive world modeling toward active embodied interaction. The architecture aims to close
- Transcript
September 13 · 13 min
Axis's Real2Sim2Real Bet: Make the Data Prove Itself
Axis introduced a Real2Sim2Real pipeline that converts real objects and scenes into high-fidelity physics-accurate simulations and validates learned policies back on hardware, paired with a performance-driven active trajectory curation engine. This data f
- Transcript
September 12 · 30 min
A Rubik's Cube, Two Robot Hands, and One Category Error
GPT-6 Astra is demonstrated solving a Rubik's Cube using dexterous robot hands in a real physical setup, showcasing a significant leap in vision-language-action integration for fine-grained dexterous manipulation. This demo highlights the frontier of embo
- Transcript
September 12 · 33 min
UnifoLM: Predict the Interaction, Not the Whole World
Unitree's UnifoLM is a 6B-parameter world model trained on 2,500 hours of real robot data that predicts only dynamic pixels and optical flow rather than full frames, enabling efficient robot action prediction. This interaction-centric design reduces compu
- Transcript
September 11 · 32 min
Stop Making the Video Model Remember the World
A new world model architecture that decouples world-state evolution from rendering: an agent writes executable rules, a persistent engine tracks entities, and a video model renders the output. This separation enables more controllable and interpretable wo
- Transcript
September 11 · 13 min
Atlas Turns Phone Scans Into Robot Worlds—But Not Yet Into Physics
Atlas is a unified spatial intelligence model that reconstructs real scenes from 2–3 images, generates 1440p video along camera paths, and supports real-to-sim robotics workflows using Gaussian Splats to create photorealistic 3D digital twins from phone s
- Transcript

Showing 61–80 of 110 episodes