Embodied AI 101 cover art

All Episodes

Embodied AI 101 — 44 episodes

#
Title
1

Claw-Eval: Toward Trustworthy and Transparent Evaluation of Autonomous Agents

2

LIBERO-Para: Paraphrase Robustness in Robotic Manipulation

3

YOR: Your Own Mobile Manipulator for Generalizable Robotics

4

EgoSim: Egocentric World Simulator for Embodied Interaction Generation

5

Accelerating Video World Models: From Generative Videos to Real-Time Simulators

6

From Tokens to Thoughts: Continuous Latent Reasoning in Large Models and Robot Control

7

CaP-X: Coding Agents for Physical eXecution

8

DoRA: Weight-Decomposed Low-Rank Adaptation

9

AI Model Collapse: What Happens When AI Trains on Its Own Outputs

10

PhAIL: Benchmarking Vision-Language-Action Models on Real-World Bin-Picking

11

Co-training Large Behavior Models: Data Modalities and Training Strategies for Robot Manipulation

12

HyDRA: Hybrid Memory for Dynamic Video World Models

13

# WildWorld: Dynamic World Modeling with Actions and Explicit State

14

Omni-WorldBench: Evaluating Interactive 4D World Models

15

SIMART: From Static Meshes to Sim-Ready Articulated Models

16

EgoSim: An Egocentric World Simulator for Embodied Interaction

17

Digit's New Motor Cortex: Sim-to-Real RL for Whole-Body Control

18

EgoNav: Diffusion-Based Humanoid Navigation from Human Egocentric Video

19

CaP-X: A Code-as-Policy Framework for Robot Manipulation

20

Embodied Intelligence Breakthrough: Generalist AI’s GEN-1 Robots

21

CaP-X: LMs' First Physical Exam

22

AI Model Collapse: The Danger of Training on AI-Generated Data

23

High-Level Automated Reasoning with Qwen2.5-7B

24

Co-Training Large Behavior Models: Multimodal Data for Robot Manipulation

25

HyDRA: Hybrid Memory for Dynamic Video World Models

26

DexWM: Leveraging Human Videos for Dexterous Robot World Models

27

World Models in Robotics

28

SIMART: Decomposing Monolithic Meshes into Sim-Ready Articulated Assets

29

LeWorldModel: A Stable JEPA World Model from Pixels

30

World Models for Robots: The Next Big Leap?

31

Harnessing Long-Running AI in Embodied Systems

32

HoMMI: Learning Whole-Body Mobile Manipulation from Human Demonstrations

33

TurboQuant: Redefining AI Efficiency with Extreme Compression

34

DexWM: Learning Dexterous Object Manipulation from Human Videos

35

FlashAttention-3: Fast & Accurate Attention with Asynchrony & Low-Precision

36

When AI Trains on Its Own Output: The Model Collapse Problem

37

MolmoBot: A Vision-Language Model for Zero-Shot Robot Manipulation

38

LeWorldModel: Stable End-to-End JEPA from Pixels

39

EgoVerse: An Egocentric Data Ecosystem for Scaling Robot Learning

40

HSImul3R: Physics-Driven Reconstruction of Human–Scene Interactions

41

MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation

42

DreamZero: World Action Models Are Zero-Shot Policies

43

Kinema4D: A 4D Generative Simulator for Embodied AI

44

VEGA-3D: Teaching multimodal LLMs spatial reasoning through video generation