Jiaxu Wang
The Chinese University of Hong Kong
収録論文 20本 ・ フィジカルAI/ロボット学習
VLA
※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。
論文
- τ0-VLA: 世界モデル誘導のテスト時計算を備えた階層型ロボット基盤モデルVLA2026/8/17
長期的なロボット操作タスクを、高レベル方策がサブタスク生成時にテスト時計算を追加で行える階層型VLAモデルを提案し、実データで性能向上を実証した。
- $\tau_0$-WM: A Unified Video-Action World Model for Robotic Manipulation2026/6/1
- MoSA: Motion-constrained Stress Adaptation for Mitigating Real-to-Sim Gap in Continuum Dynamics via Learning Residual Anisotropy2026/5/1
- Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation2026/5/1
- MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction2026/2/1
- DPL: Depth-only Perceptive Humanoid Locomotion via Realistic Depth Synthesis and Cross-Attention Terrain Reconstruction2025/10/1
- Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Composition2025/10/1
- Humanoid Occupancy: Enabling A Generalized Multimodal Occupancy Perception System on Humanoid Robots2025/7/1
- Occupancy World Model for Robots2025/5/1
- Modality-Composable Diffusion Policy via Inference-Time Distribution-level Composition2025/3/1
- EmbodiedVSR: Dynamic Scene Graph-Guided Chain-of-Thought Reasoning for Visual Spatial Tasks2025/3/1
- HumanoidPano: Hybrid Spherical Panoramic-LiDAR Cross-Modal Perception for Humanoid Robots2025/3/1
- Distillation-PPO: A Novel Two-Stage Reinforcement Learning Framework for Humanoid Robot Perceptive Locomotion2025/3/1
- Trinity: A Modular Humanoid Robot AI System2025/3/1
- LiPS: Large-Scale Humanoid Robot Reinforcement Learning with Parallel-Series Structures2025/3/1
- E2H: A Two-Stage Non-Invasive Neural Signal Driven Humanoid Robotic Whole-Body Control Framework2024/10/1
- Mamba Policy: Towards Efficient 3D Diffusion Policy with Hybrid Selective State Models2024/9/1
- Query-based Semantic Gaussian Field for Scene Representation in Reinforcement Learning2024/6/1
- Physical Priors Augmented Event-Based 3D Reconstruction2024/1/1
- Chasing Day and Night: Towards Robust and Efficient All-Day Object Detection Guided by an Event Camera2023/9/1