Modeling and Understanding Neural Weights of Humanoid Policies with Diffusion
cs.RO, cs.AI, cs.LG
Submitted: 2026-10-14
Updated: 2026-10-14
Comments: Project page: https://humanoid-weight-diffusion.github.io/
Project page: https://humanoid-weight-diffusion.github.io
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- HumanPlus: Humanoid Shadowing and Imitation from Humans
- OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning
- ExBody2: Advanced Expressive Humanoid Whole-Body Control
- BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion
- SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control
- HyperNetworks
- Neural Network Diffusion
- DeepWeightFlow: Re-Basined Flow Matching for Generating Neural Network Weights
- Generative Modeling of Weights: Generalization or Memorization?
- Generating Behaviorally Diverse Policies with Latent Diffusion Models
- WARPD: World model Assisted Reactive Policy Diffusion
- Does the Data Induce Capacity Control in Deep Learning?
- The Low-Rank Simplicity Bias in Deep Networks
- Sloppiness and Emergent Theories in Physics, Biology, and Beyond
- On Lottery Tickets and Minimal Task Representations in Deep Reinforcement Learning
- mjlab: A Lightweight Framework for GPU-Accelerated Robot Learning
- Back to Basics: Let Denoising Generative Models Denoise
- Git Re-Basin: Merging Models modulo Permutation Symmetries
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies