Yu-Gang Jiang

Fudan University

収録論文 41本 ・ フィジカルAI/ロボット学習

※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

• StructRL: フローベースVLAのための構造化アクション空間探索
  VLA/強化学習2026/8/15

  フローベースの視覚言語行動モデル(VLA)のオンライン強化学習において、ノイズをアクション空間に直接注入する構造化探索手法StructRLを提案し、シミュレーションと実世界タスクで性能を向上させた。

• HandEdit: 身体性を考慮した人間からロボットへの器用な手の画像編集のための統一ベンチマーク
  画像編集/ロボティクス2026/8/12

  人間の手の画像を様々なロボットハンドに変換する大規模な画像編集データセットとベンチマークを構築し、既存の編集モデルの評価と身体性を考慮した編集モデルの発展を促進する。

• ETA: 身体性タスクのための新しいエージェントパラダイム
  エージェントアーキテクチャ2026/8/4

  ロボットのChatGPT的瞬間を目指し、プランナー・インターフェース・ワールドのループで構成される新しい身体性タスクエージェント（ETA）を提案し、オープンソース実装OpenETAを公開した。

• HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing
  2026/8/1
• ETA: A New Agentic Paradigm for Embodied Tasks
  2026/8/1
• EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstration
  2026/7/1
• SegDiff: Segmented Trajectory Diffusion for Consistent and Adaptive Robot Manipulation
  2026/7/1
• Coarse-to-Control: Action-Token Planning for Vision-Language-Action Models
  2026/6/1
• ActiveMimic: Egocentric Video Pretraining with Active Perception
  2026/6/1
• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data
  2026/6/1
• UniDexTok: A Unified Dexterous Hand Tokenizer from Real Data
  2026/6/1
• Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlation
  2026/6/1
• Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy
  2026/6/1
• ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation
  2026/6/1
• Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations
  2026/5/1
• GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization
  2026/5/1
• VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models
  2026/5/1
• Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
  2026/5/1
• Afford-VLA: Action-Aligned Visual Planning via Internalized Affordance
  2026/5/1
• World Action Models: The Next Frontier in Embodied AI
  2026/5/1
• HazardArena: Evaluating Semantic Safety in Vision-Language-Action Models
  2026/4/1
• AssemLM: A Spatial Reasoning Multimodal Large Language Model for Robotic Assembly
  2026/4/1
• Robotic Grasping and Placement Controlled by EEG-Based Hybrid Visual and Motor Imagery
  2026/3/1
• OCRA: Object-Centric Learning with 3D and Tactile Priors for Human-to-Robot Action Transfer
  2026/3/1
• RC-NF: Robot-Conditioned Normalizing Flow for Real-Time Anomaly Detection in Robotic Manipulation
  2026/3/1
• FRoM-W1: Towards General Humanoid Whole-Body Control with Language Instructions
  2026/1/1
• Schr\"odinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation
  2025/12/1
• HiMoE-VLA: Hierarchical Mixture-of-Experts for Generalist Vision-Language-Action Policies
  2025/12/1
• Unify Robot Actions in Camera Frame
  2025/11/1
• TP-MDDN: Task-Preferenced Multi-Demand-Driven Navigation with Autonomous Decision-Making
  2025/11/1
• RoboOmni: Proactive Robot Manipulation in Omni-modal Context
  2025/10/1
• DropVLA: An Action-Level Backdoor Attack on Vision-Language-Action Models
  2025/10/1
• Ask-to-Clarify: Resolving Instruction Ambiguity through Multi-turn Dialogue
  2025/9/1
• Embodied AI: From LLMs to World Models
  2025/9/1
• Actor-Critic for Continuous Action Chunks: A Reinforcement Learning Framework for Long-Horizon Robotic Manipulation with Sparse Reward
  2025/8/1
• HumanoidGen: Data Generation for Bimanual Dexterous Manipulation via LLM Reasoning
  2025/7/1
• TriVLA: A Triple-System-Based Unified Vision-Language-Action Model with Episodic World Modeling for General Robot Control
  2025/7/1
• You Only Estimate Once: Unified, One-stage, Real-Time Category-level Articulated Object 6D Pose Estimation for Robotic Grasping
  2025/6/1
• Human2Robot: Learning Robot Actions from Paired Human-Robot Videos
  2025/2/1
• SparseGrasp: Robotic Grasping via 3D Semantic Gaussian Splatting from Sparse Multi-View RGB Images
  2024/12/1
• VLABench: A Large-Scale Benchmark for Language-Conditioned Robotics Manipulation with Long-Horizon Reasoning Tasks
  2024/12/1