10 papers
MuSTMM: Learning Sound-Triggered Mobile Manipulation across Multiple Scenes
Hao Ju, Feng Yan, Zhihao Zhong +3
Sound-Triggered Mobile Manipulation (STMM) requires robots to respond to sound events by navigating to the associated objects and performing appropriate physical interactions. Exis…
Can a Frontier VLM Serve as a Robot Policy for Memory-Dependent Manipulation?
Jack B. Jedlicki, Amish Sethi, Heng Yang
We use a frozen frontier vision-language model, GPT-6-Astra, directly as the robot policy on RoboMME, a benchmark of 16 memory-dependent manipulation tasks spanning temporal, spati…
Where Memory Belongs: Ledger, an Object Ledger for Memory-Augmented VLAs
Tanguy Dieudonné, Jack B. Jedlicki, Heng Yang
Memory is essential for long-horizon, partially observed robotic manipulation: a robot must remember which object was placed in a drawer, whose cup it moved, or how many action cyc…
RoboMP-DINOv2: Prompts, Not Filters for Robust Robot Manipulation
Han Qi, Heng Yang
Robot manipulation policies must generalize across visual shifts while preserving scene context relevant to action. General-purpose vision encoders are not tailored to visuomotor c…
Towards Safe Mobility: A Unified Transportation Foundation Model enabled by Open-Ended Vision-Language Dataset
Wenhui Huang, Songyan Zhang, Collister Chua +4
Urban transportation systems face growing safety challenges that require scalable intelligence for emerging smart mobility infrastructures. While recent advances in foundation mode…
Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
Rishabh Tiwari, Aditya Tomar, Udbhav Bamba +5
Process Reward Models (PRMs) are rapidly becoming the backbone of LLM reasoning pipelines, yet we demonstrate that state-of-the-art PRMs are systematically exploitable under advers…
MoTVLA: A Vision-Language-Action Model with Unified Fast-Slow Reasoning
Wenhui Huang, Changhe Chen, Han Qi +3
Integrating visual-language instructions into visuomotor policies is gaining momentum in robot learning for enhancing open-world generalization. Despite promising advances, existin…
Compose by Focus: Scene Graph-based Atomic Skills
Han Qi, Changhe Chen, Heng Yang
A key requirement for generalist robots is compositional generalization - the ability to combine atomic skills to solve complex, long-horizon tasks. While prior work has primarily…
Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling
Han Qi, Haocheng Yin, Aris Zhu +2
We present Generative Predictive Control (GPC), an inference-time method for improving pretrained behavior-cloning policies without retraining. GPC augments a frozen diffusion poli…
Control-oriented Clustering of Visual Latent Representation
Han Qi, Haocheng Yin, Heng Yang
We initiate a study of the geometry of the visual representation space -- the information channel from the vision encoder to the action decoder -- in an image-based control pipelin…