Source author record

Xiaomeng Xu

Xiaomeng Xu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

math-ph math.MP math.DG math.CA math.SG Robotics Computer Vision hep-th math.AP math.CT math.QA

Catalog footprint

What is connected

9works

11topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

In-the-Wild Compliant Manipulation with UMI-FT

Many manipulation tasks require careful force modulation. With insufficient force the task may fail, while excessive force could cause damage. The high cost, bulky size and fragility of commercial force/torque (F/T) sensors have limited large-scale, force-aware policy learning. We introduce UMI-FT, a handheld data-collection platform that mounts compact, six-axis force/torque sensors on each finger, enabling finger-level wrench measurements alongside RGB, depth, and pose. Using the multimodal data collected from this device, we train an adaptive compliance policy that predicts position targets, grasp force, and stiffness for execution on standard compliance controllers. In evaluations on three contact-rich, force-sensitive tasks (whiteboard wiping, skewering zucchini, and lightbulb insertion), UMI-FT enables policies that reliably regulate external contact forces and internal grasp forces, outperforming baselines that lack compliance or force sensing. UMI-FT offers a scalable path to learning compliant manipulation from in-the-wild demonstrations. We open-source the hardware and software to facilitate broader adoption at:https://umi-ft.github.io/.

preprint2026arXiv

RoboPanoptes: The All-seeing Robot with Whole-body Dexterity

We present RoboPanoptes, a capable yet practical robot system that achieves whole-body dexterity through whole-body vision. Its whole-body dexterity allows the robot to utilize its entire body surface for manipulation, such as leveraging multiple contact points or navigating constrained spaces. Meanwhile, whole-body vision uses a camera system distributed over the robot's surface to provide comprehensive, multi-perspective visual feedback of its own and the environment's state. At its core, RoboPanoptes uses a whole-body visuomotor policy that learns complex manipulation skills directly from human demonstrations, efficiently aggregating information from the distributed cameras while maintaining resilience to sensor failures. Together, these design aspects unlock new capabilities and tasks, allowing RoboPanoptes to unbox in narrow spaces, sweep multiple or oversized objects, and succeed in multi-step stowing in cluttered environments, outperforming baselines in adaptability and efficiency. Results are best viewed on https://robopanoptes.github.io.

preprint2022arXiv

AutoGPart: Intermediate Supervision Search for Generalizable 3D Part Segmentation

Training a generalizable 3D part segmentation network is quite challenging but of great importance in real-world applications. To tackle this problem, some works design task-specific solutions by translating human understanding of the task to machine's learning process, which faces the risk of missing the optimal strategy since machines do not necessarily understand in the exact human way. Others try to use conventional task-agnostic approaches designed for domain generalization problems with no task prior knowledge considered. To solve the above issues, we propose AutoGPart, a generic method enabling training generalizable 3D part segmentation networks with the task prior considered. AutoGPart builds a supervision space with geometric prior knowledge encoded, and lets the machine to search for the optimal supervisions from the space for a specific segmentation task automatically. Extensive experiments on three generalizable 3D part segmentation tasks are conducted to demonstrate the effectiveness and versatility of AutoGPart. We demonstrate that the performance of segmentation networks using simple backbones can be significantly improved when trained with supervisions searched by our method.

preprint2022arXiv

On the connection formula of a higher rank analog of Painlevé VI

In this paper, we derive the connection formula for the higher rank analog of Painlevé VI arising from the isomonodromic deformation theory and the study of Frobenius manifolds.

preprint2020arXiv

Frobenius manifolds and quantum groups

We introduce isomonodromy Knizhnik-Zamolodchikov (KZ) connections with respect to the quantum Stokes matrices, and prove that the classical limit of the KZ type connections gives rise to the Dubrovin connections of semisimple Frobenius manifolds and flat F-manifolds.

preprint2016arXiv

The Pontryagin Class for Pre-Courant Algebroids

In this paper, we show that the Jacobiator $J$ of a pre-Courant algebroid is closed naturally. The corresponding equivalence class $[J^\flat]$ is defined as the Pontryagin class, which is the obstruction of a pre-Courant algebroid to be deformed into a Courant algebroid. We construct a Leibniz 2-algebra and a Lie 2-algebra associated to a pre-Courant algebroid and prove that these algebraic structures are isomorphic under deformations. Finally, we introduce the twisted action of a Lie algebra on a manifold to give more examples of pre-Courant algebroids, which include the Cartan geometry.

preprint2014arXiv

Canonical Functions and Differential Graded Symplectic Pairs in Supergeometry and AKSZ Sigma Models with Boundary

Consistent boundary conditions for Alexandrov-Kontsevich-Schwartz-Zaboronsky (AKSZ) sigma models and the corresponding boundary theories are analyzed. As their mathematical structures, we introduce a generalization of differential graded symplectic manifolds, called twisted QP manifolds, in terms of graded symplectic geometry, canonical functions, and QP pairs. We generalize the AKSZ construction of topological sigma models to sigma models with Wess-Zumino terms and show that all the twisted Poisson-like structures known in the literature can actually be naturally realized as boundary conditions for AKSZ sigma models.

preprint2012arXiv

Cohomology for almost Lie algebroids

We give numerous examples of almost Lie algebroids arising as Dirac structures in pre-Courant algebroids, e.g. from twisted Poisson structures, as well as from twisted actions of a Lie algebra. We moreover define a cohomology for them, motivated by a Q-structure, that is trivial for (bundles of) Lie algebras but characterizes the underlying integrable distribution the image of the anchor map.