← Tags
#机器人操作 (18)
τ0-VLA: Robot VLA that searches over subtask sequences at test time with a world model
VLA世界模型World Model测试时搜索机器人操作
Riemann-1.0 trains a World Action Model on 200k hours of human video
World Action Model世界动作模型具身智能Embodied AI人类视频
Splitting Robot Reaching into Alignment and Interaction Phases
机器人操作manipulation抓取策略学习两阶段学习
Patch Policy: small transformer outperforms large VLAs
Patch PolicyVLAtransformer机器人操作NYU
ADEPT: Sim Pre-training for Zero-Shot Visuo-Tactile Dexterity
ADEPT灵巧手灵巧操作视触觉触觉传感
HiFi-UMI: Learning Manipulation Policies from High-Fidelity UMI Data Alone
操作策略manipulation policyUMIHiFi-UMI通用操作接口
Long-Horizon Memory in Robotics Without Policy Retraining
长时序记忆机器人操作策略学习记忆系统long-horizon memory
Patch Policy: Efficient Embodied Control via Dense Visual Representations
Patch Policydense visual representationDINOv2WebSSLVLA
FlashVLA: Streaming Action Decoding for Fast and Asynchronous VLA Inference
VLAflow matching流式解码异步推理低延迟
PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable Dynamics
世界模型物理仿真可变形物体机器人操作Paper
AXIS: A Growable Community-Driven Data Engine for Scalable Robot Manipulation
机器人操作数据引擎社区驱动模仿学习Paper
Addressing the Orchestration Gap in Generalist Robots via Physical Agency
TwitterVLA机器人操作编排具身AI
Why Does Action Chunking Improve Behavioral Cloning Performance in Robotic Control?
动作分块行为克隆机器人操作隐式集成扩散策略
Ego2Robot: Scalable Robot Data Synthesis from Egocentric Human Data
VLA机器人操作数据合成第一人称视频动作重定向
StageWAM: Joint-Embedding Stage Prediction for World-Action Models in Robot Manipulation
机器人操作世界模型JEPAWAMVLA
Scale Up Strategically: Learning Compositional Generalization via Bias-Aware Evaluation and Data Collection for Robotic Manipulation
机器人操作组合泛化数据采集偏置Paper
ViTacWorld: Scaling Visuo-Tactile World Models for Contact-Rich Robot Manipulation
世界模型World Model视触觉Visuo-Tactile机器人操作
Riemann-1.0: An Embodied World Action Model for Physical AI
具身智能World Action Model因果自回归渐进式预训练Riemann Dynamics