syksykCCC

一些关于 3D Robotic 和 Vision-Language-Action (VLA) 大模型的论文选读


3D-MVP: 3D Multiview Pretraining for Robotic Manipulation [arxiv] [note]

3D Diffuser Actor: Policy Diffusion with 3D Scene Representations [arxiv] [note]

3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations [arxiv] [note]

Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding [arxiv] [note]

Act3D: 3D Feature Field Transformers for Multi-Task Robotic Manipulation [arxiv] [note]

Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders [arxiv] [note]

Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning [arxiv] [note]

PolarNet: 3D Point Clouds for Language-Guided Robotic Manipulation [arxiv] [note]

ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation [arxiv] [note]

SUGAR: Pre-training 3D Visual Representations for Robotics [arxiv] [note]

SPA: 3D Spatial-Awareness Enables Effective Embodied Representation [arxiv] [note]

RVT: Robotic View Transformer for 3D Object Manipulation [arxiv] [note]

RVT-2: Learning Precise Manipulation from Few Demonstrations [arxiv] [note]


3D-VLA: A 3D Vision-Language-Action Generative World Model [arxiv] [note]

AffordDP: Generalizable Diffusion Policy with Transferable Affordance [arxiv] [note]

Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding [arxiv] [note]

CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation [arxiv] [note]

DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution [arxiv] [note]

DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control [arxiv] [note]

FAST: Efficient Action Tokenization for Vision-Language-Action Models [arxiv] [note]

Improving Vision-Language-Action Models via Chain-of-Affordance [arxiv] [note]

Diffusion-VLA: Scaling Robot Foundation Models via Unified Diffusion and Autoregression [arxiv] [note]

RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation [arxiv] [note]

Robo-ABC: Affordance Generalization Beyond Categories via Semantic Correspondence for Robot Manipulation [arxiv] [note]