19 papers · 1 filter
How High Can GPT-6 Astra Climb in Contact-Rich Manipulation? PhysLadder: A Hierarchical, Matched Sim-Real Benchmark for Physical Interaction
Haibao Yu, Jiaqi Meng, Zijian Zhang +16
We introduce PhysLadder, a hierarchical Sim-Real benchmark for evaluating physical interaction capabilities in contact-rich manipulation. Its 16 household tasks span three cumulati…
PRICE the Action Chunks: Physical Relational Credit Assignment for Embodied Reinforcement Learning
Yangang Zou, Jiajun Lu, Weitao Zhou +6
Outcome-based reinforcement learning (RL) post-trains vision--language--action policies using terminal success signals, but assigns the same trajectory-level advantage to every act…
GaussianDream++: Efficient 3D Gaussian World Modeling for Robotic Manipulation
Yuqing Jiang, Zijian Zhang, Weitao Zhou +8
Vision-Language-Action (VLA) policies have advanced language-conditioned robotic manipulation, yet action-imitation objectives provide only weak supervision for metric 3D structure…
GaussianWAM: Distilling Geometry and Semantics from 3D Gaussian Fields into World-Action Models
Zijian Zhang, Yuqing Jiang, Weitao Zhou +6
World-Action Models (WAMs) jointly learn future visual prediction and action generation, using video dynamics as a representation-learning signal for robotic manipulation. However,…
SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation
Bowen Jing, Mingxin Wang, Ruiyang Hao +15
Physical interaction quality is central to deformable-object manipulation, yet most benchmarks evaluate task success alone. A policy may complete the task while allowing slip or ca…
SoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects (Early Version)
Bowen Jing, Mingxin Wang, Ruiyang Hao +15
Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or…
HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone
Simple AI, :, Yuteng Wei +16
Learning deployable manipulation policies is bottlenecked by the scarcity of data that is both high-fidelity and scalable. Real-robot teleoperation is accurate but costly to scale;…
EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstration
Zuhao Ge, Yuchen Zhou, Weitao Zhou +8
Robust embodied robots should be able to recover from failures and retry tasks in order to operate reliably in unstructured and noisy real-world environments. Achieving this capabi…
Diagnosing Semantic Handoff Failures in Agent-Orchestrated Vision-Language-Action Skill Composition
Ke Rui, Yushen Zuo, Jiawei Wang +4
Long-horizon household tasks require robots to compose many language-conditioned skills, yet the boundary between consecutive skills is rarely explicit. A skill may satisfy its own…
Dynamics Are Learned, Not Told: Semi-Supervised Discovery of Latent Dynamics Geometries For Zero-Shot Policy Adaptation
Zhiming Xu, Weitao Zhou, Xianghui Pan +4
Real-world dynamics shifts pose a critical challenge for reinforcement learning in robotics, as policies tightly coupled to nominal environments often fail catastrophically when ph…
GaussianDream: A Feed-Forward 3D Gaussian World Model for Robotic Manipulation
Zijian Zhang, Yuqing Jiang, Qian Cheng +6
Vision-language-action (VLA) policies have advanced language-conditioned robotic manipulation by transferring semantic priors from pretrained vision-language models to action gener…
CounterScene: Counterfactual Causal Reasoning in Generative World Models for Safety-Critical Closed-Loop Evaluation
Bowen Jing, Ruiyang Hao, Weitao Zhou +1
Generating safety-critical driving scenarios requires understanding why dangerous interactions arise, rather than merely forcing collisions. However, existing methods rely on heuri…
LSRE: Latent Semantic Rule Encoding for Real-Time Semantic Risk Detection in Autonomous Driving
Qian Cheng, Weitao Zhou, Cheng Jing +5
Real-world autonomous driving must adhere to complex human social rules that extend beyond legally codified traffic regulations. Many of these semantic constraints, such as yieldin…
Are All Data Necessary? Efficient Data Pruning for Large-scale Autonomous Driving Dataset via Trajectory Entropy Maximization
Zhaoyang Liu, Weitao Zhou, Junze Wen +4
Collecting large-scale naturalistic driving data is essential for training robust autonomous driving planners. However, real-world datasets often contain a substantial amount of re…
DTCCL: Disengagement-Triggered Contrastive Continual Learning for Autonomous Bus Planners
Yanding Yang, Weitao Zhou, Jinhai Wang +8
Autonomous buses run on fixed routes but must operate in open, dynamic urban environments. Disengagement events on these routes are often geographically concentrated and typically…
Sequence of Expert: Boosting Imitation Planners for Autonomous Driving through Temporal Alternation
Xiang Li, Gang Liu, Weitao Zhou +2
Imitation learning (IL) has emerged as a central paradigm in autonomous driving. While IL excels in matching expert behavior in open-loop settings by minimizing per-step prediction…
DRARL: Disengagement-Reason-Augmented Reinforcement Learning for Efficient Improvement of Autonomous Driving Policy
Weitao Zhou, Bo Zhang, Zhong Cao +5
With the increasing presence of automated vehicles on open roads under driver supervision, disengagement cases are becoming more prevalent. While some data-driven planning systems…
Dynamically Local-Enhancement Planner for Large-Scale Autonomous Driving
Nanshan Deng, Weitao Zhou, Bo Zhang +4
Current autonomous vehicles operate primarily within limited regions, but there is increasing demand for broader applications. However, as models scale, their limited capacity beco…
Dynamically Conservative Self-Driving Planner for Long-Tail Cases
Weitao Zhou, Zhong Cao, Nanshan Deng +3
Self-driving vehicles (SDVs) are becoming reality but still suffer from "long-tail" challenges during natural driving: the SDVs will continually encounter rare, safety-critical cas…