VLA Foundry: A Unified Framework for Training Vision-Language-Action Models

TL;DR: A unified open-source training framework for vision-language-action models, consolidating LLM, VLM, and VLA training into a single modular codebase.

A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation

TL;DR: Carefully study the performance of LBMs to characterize their performance and scaling.

CodeDiffuser: Attention-Enhanced Diffusion Policy via VLM-Generated Code for Instruction Ambiguity

TL;DR: Use a VLM to generate attention maps from language inputs that inform a 3D diffusion policy.

OmniShape: Zero-Shot Multi-Hypothesis Shape and Pose Estimation in the Real World

TL;DR: Decompose shape and pose estimation into two stages, both modelled by diffusion models.

ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping

TL;DR: A zero-shot robotic grasping system that predicts completed 3D object shapes and grasps.

Zero-Shot Multi-object Scene Completion

SE(3) Equivariant Ray Embeddings for Implicit Multi-View Depth Estimation

View-Invariant Policy Learning via Zero-Shot Novel View Synthesis

ReFiNe: Recursive Field Networks for Cross-Modal Multi-Scene Representation

TL;DR: A recursive neural field architecture that efficiently represents and renders multiple scenes across different modalities like RGB, depth, and semantics.

DiffusionNOCS: Managing Symmetry and Uncertainty in Sim2Real Multi-Modal Category-level Pose Estimation

TL;DR: Diffuse NOCS images to generate category-level object poses.

NeO 360: Neural Fields for Sparse View Synthesis of Outdoor Scenes

CARTO: Category and Joint Agnostic Reconstruction of ARTiculated Objects

Multi-Object Manipulation via Object-Centric Neural Scattering Functions

ROAD: Learning an Implicit Recursive Octree Auto-Decoder to Efficiently Encode 3D Shapes

VoluMon: Weakly-Supervised Volumetric Monocular Estimation with Ellipsoid Representations

Hierarchical Object Map Estimation for Efficient and Robust Navigation

Online High-Level Model Estimation for Efficient Hierarchical Robot Navigation

Search and Rescue Under the Forest Canopy Using Multiple UAVs

Learned Sampling Distributions for Efficient Planning in Hybrid Geometric and Object-Level Representations

Robust Object-Based SLAM for High-Speed Autonomous Navigation

Deep Inference for Covariance Estimation: Learning Gaussian Noise Models for State Estimation