arXiv subjects

Jian Tang

Publications and source records attributed to Jian Tang.

At least 109 records · Page 6Linked to original sources

arXiv2025

Interaction-Driven Topological Transitions in Monolayer TaIrTe$_4$

Discovering materials that combine topological phenomena with correlated electron behavior is a central pursuit in quantum materials research. Monolayer TaIrTe$_4$ has recently emerged as a promising platform in this context, hosting robust quantum spin Hall insulator (QSHI) phases both within a single-particle gap and within a correlation-induced gap arising from van Hove singularities (vHSs), accessed via electrostatic doping. Its intrinsic monolayer nature offers exceptional tunability and the potential to realize a versatile array of interaction-driven topological phases. In this work, we combine theory and experiment to map the phase landscape of monolayer TaIrTe$_4$. Using Hartree-Fock calculations, we investigate the interaction-driven phase diagram near the vHSs under commensurate filling conditions. By systematically tuning the dielectric screening and strain, we uncover a rich set of ground states--including QSHI, trivial insulator, higher-order topological insulator, and metallic phase--among which are interaction-driven topological phase transitions. Experimentally, we perform both local and nonlocal transport measurements across a broad set of devices, which--due to unavoidable strain variations during fabrication-realize several phases consistent with theoretical predictions. Together, our results lay the groundwork for understanding correlation-driven topological phenomena in TaIrTe$_4$ and open new directions for engineering exotic quantum phases in low-dimensional materials beyond the limitations of moiré superlattices.

cond-mat.str-el↗

arXiv2025

SwitchVLA: Execution-Aware Task Switching for Vision-Language-Action Models

Robots deployed in dynamic environments must be able to not only follow diverse language instructions but flexibly adapt when user intent changes mid-execution. While recent Vision-Language-Action (VLA) models have advanced multi-task learning and instruction following, they typically assume static task intent, failing to respond when new instructions arrive during ongoing execution. This limitation hinders natural and robust interaction in dynamic settings, such as retail or household environments, where real-time intent changes are common. We propose SwitchVLA, a unified, execution-aware framework that enables smooth and reactive task switching without external planners or additional switch-specific data. We model task switching as a behavior modulation problem conditioned on execution state and instruction context. Expert demonstrations are segmented into temporally grounded contact phases, allowing the policy to infer task progress and adjust its behavior accordingly. A multi-behavior conditional policy is then trained to generate flexible action chunks under varying behavior modes through conditioned trajectory modeling. Experiments in both simulation and real-world robotic manipulation demonstrate that SwitchVLA enables robust instruction adherence, fluid task switching, and strong generalization-outperforming prior VLA baselines in both task success rate and interaction naturalness.

cs.RO↗

arXiv2025

Aligning Protein Conformation Ensemble Generation with Physical Feedback

Protein dynamics play a crucial role in protein biological functions and properties, and their traditional study typically relies on time-consuming molecular dynamics (MD) simulations conducted in silico. Recent advances in generative modeling, particularly denoising diffusion models, have enabled efficient accurate protein structure prediction and conformation sampling by learning distributions over crystallographic structures. However, effectively integrating physical supervision into these data-driven approaches remains challenging, as standard energy-based objectives often lead to intractable optimization. In this paper, we introduce Energy-based Alignment (EBA), a method that aligns generative models with feedback from physical models, efficiently calibrating them to appropriately balance conformational states based on their energy differences. Experimental results on the MD ensemble benchmark demonstrate that EBA achieves state-of-the-art performance in generating high-quality protein ensembles. By improving the physical plausibility of generated structures, our approach enhances model predictions and holds promise for applications in structural biology and drug discovery.

q-bio.BM↗

arXiv2025

RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulation

In this paper, we introduce RoboMIND (Multi-embodiment Intelligence Normative Data for Robot Manipulation), a dataset containing 107k demonstration trajectories across 479 diverse tasks involving 96 object classes. RoboMIND is collected through human teleoperation and encompasses comprehensive robotic-related information, including multi-view observations, proprioceptive robot state information, and linguistic task descriptions. To ensure data consistency and reliability for imitation learning, RoboMIND is built on a unified data collection platform and a standardized protocol, covering four distinct robotic embodiments: the Franka Emika Panda, the UR5e, the AgileX dual-arm robot, and a humanoid robot with dual dexterous hands. Our dataset also includes 5k real-world failure demonstrations, each accompanied by detailed causes, enabling failure reflection and correction during policy learning. Additionally, we created a digital twin environment in the Isaac Sim simulator, replicating the real-world tasks and assets, which facilitates the low-cost collection of additional training data and enables efficient evaluation. To demonstrate the quality and diversity of our dataset, we conducted extensive experiments using various imitation learning methods for single-task settings and state-of-the-art Vision-Language-Action (VLA) models for multi-task scenarios. By leveraging RoboMIND, the VLA models achieved high manipulation success rates and demonstrated strong generalization capabilities. To the best of our knowledge, RoboMIND is the largest multi-embodiment teleoperation dataset collected on a unified platform, providing large-scale and high-quality robotic training data. Our project is at https://x-humanoid-robomind.github.io/.

cs.RO↗

arXiv2025

MuGrid-v2: A novel scintillator detector for multidisciplinary applications

Muography, traditionally recognized as a potent instrument for imaging the internal structure of gigantic objects, has initialized various interdisciplinary applications. As the financial and labor costs of muography detector development hinder their massive applications, we develop a novel muon detector called MuGrid by coupling a monolithic plastic scintillator with the light guide array in order to achieve competitive spatial resolution while substantially reducing production costs. For a prototype detector in 30 cm $\times$ 30 cm, the intrinsic spatial resolution has been optimized toward a millimeter scale. An outdoor field muography experiment was conducted to monitor two buildings for validation purposes. The test successfully resolved the geometric influence of architectural features based on the attenuation of muon flux in a good agreement between experimental results and the simulation prediction.

physics.ins-det↗

arXiv2025

Quantum spin Hall effects in van der Waals materials

The quantum spin Hall (QSH) effect, first predicted in graphene by Kane and Mele in 2004, has emerged as a prototypical platform for exploring spin-orbit coupling, topology, and electronic interactions. Initially realized experimentally in quantum wells exhibiting characteristic QSH signatures, the field has since expanded with the discovery of van der Waals (vdW) materials. This review focuses on vdW systems, which offer unique advantages: their exposed surfaces enable a combination of surface-sensitive spectroscopic and microscopic tools for comprehensive detection of the QSH state; mechanical stacking with other vdW layers facilitates symmetry engineering and proximity effects; and moiré engineering introduces layer skyrmion topological phases and strong correlation effects. We highlight two monolayer families, 1T$^\prime$-MX$_2$ and MM$^\prime$X$_4$, represented by WTe$_2$ and TaIrTe$_4$, respectively. These materials exhibit QSH phases intertwined with or in close proximity to other quantum phases, such as excitonic insulators, charge density waves, and superconductivity. Their low crystal symmetry and topology enable rich quantum geometrical responses, ranging from nonlinear Hall effects to circular photogalvanic effects. We also discuss moiré systems, which combine topology with flatband physics and enhanced correlations, driving spontaneous symmetry breaking and transitions from QSH to quantum anomalous Hall (QAH) states. Remarkably, fractionalized QAH and QSH states have recently been observed in moiré systems, significantly advancing the field of condensed matter physics. Finally, we explore emerging applications of QSH and derived materials, such as using nonlinear Hall effects for quantum rectification in microwave energy harvesting and harnessing fractional anomalous states for topological quantum computing.

cond-mat.mes-hall↗

arXiv2025

Anisotropic Lorentz invariance violation in reactor neutrino experiments

Recent reactor neutrino oscillation experiments reported precision measurements of $\sin^2 2θ_{13}$ and $Δm^2_{ee}$ under the standard 3$ν$ oscillation framework. However, inter-experiment consistency checks through the parameter goodness-of-fit test reveal proximity to the tension boundary, with the Double Chooz, RENO, and Daya Bay ensemble yielding $p_\text{PG}=0.14$ vs threshold $α=0.1$. Anisotropic Lorentz invariance violation (LIV) can accommodate this tension by introducing a location-dependent angle $θ^\text{LIV}$ relative to the earth's axis. It is found that anisotropic LIV improve the fit to data up to 1.9$σ$ confidence level significance, with the coefficient $\mathcal{A}^{C(0)}_{31} = 2.43\times 10^{-17}\text{ MeV}$ yielding the best fit, while Parameter Goodness-of-fit (PG) is significantly improved within the LIV formalism.

hep-ph↗

arXiv2025

Pantheon: Personalized Multi-objective Ensemble Sort via Iterative Pareto Policy Optimization

In this paper, we provide our milestone ensemble sort work and the first-hand practical experience, Pantheon, which transforms ensemble sorting from a "human-curated art" to a "machine-optimized science". Compared with formulation-based ensemble sort, our Pantheon has the following advantages: (1) Personalized Joint Training: our Pantheon is jointly trained with the real-time ranking model, which could capture ever-changing user personalized interests accurately. (2) Representation inheritance: instead of the highly compressed Pxtrs, our Pantheon utilizes the fine-grained hidden-states as model input, which could benefit from the Ranking model to enhance our model complexity. Meanwhile, to reach a balanced multi-objective ensemble sort, we further devise an \textbf{iterative Pareto policy optimization} (IPPO) strategy to consider the multiple objectives at the same time. To our knowledge, this paper is the first work to replace the entire formulation-based ensemble sort in industry RecSys, which was fully deployed at Kuaishou live-streaming services, serving 400 Million users daily.

cs.SI↗

arXiv2025

TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation

Vision-Language-Action (VLA) models have shown remarkable potential in visuomotor control and instruction comprehension through end-to-end learning processes. However, current VLA models face significant challenges: they are slow during inference and require extensive pre-training on large amounts of robotic data, making real-world deployment difficult. In this paper, we introduce a new family of compact vision-language-action models, called TinyVLA, which offers two key advantages over existing VLA models: (1) faster inference speeds, and (2) improved data efficiency, eliminating the need for pre-training stage. Our framework incorporates two essential components to build TinyVLA: (1) initializing the policy backbone with robust, high-speed multimodal models, and (2) integrating a diffusion policy decoder during fine-tuning to enable precise robot actions. We conducted extensive evaluations of TinyVLA in both simulation and on real robots, demonstrating that our approach significantly outperforms the state-of-the-art VLA model, OpenVLA, in terms of speed and data efficiency, while delivering comparable or superior performance. Additionally, TinyVLA exhibits strong generalization capabilities across various dimensions, including language instructions, novel objects, unseen positions, changes in object appearance, background variations, and environmental shifts, often matching or exceeding the performance of OpenVLA. We believe that \methodname offers an interesting perspective on utilizing pre-trained multimodal models for policy learning. Our project is at https://tiny-vla.github.io.

cs.RO↗

arXiv2025

Magnetization Dependent In-plane Anomalous Hall Effect in a Low-dimensional System

Anomalous Hall Effect (AHE) response in magnetic systems is typically proportional to an out-of-plane magnetization component because of the restriction imposed by system symmetries, which demands that the magnetization, applied electric field, and induced Hall current are mutually orthogonal to each other. Here, we report experimental realization of an unconventional form of AHE in a low-dimensional heterostructure, wherein the Hall response is not only proportional to the out-of-plane magnetization component but also to the in-plane magnetization component. By interfacing a low-symmetry topological semimetal (TaIrTe4) with the ferromagnetic insulator (Cr2Ge2Te6), we create a low-dimensional magnetic system, where only one mirror symmetry is preserved. We show that as long as the magnetization has a finite component in the mirror plane, this last mirror symmetry is broken, allowing the emergence of an AHE signal proportional to in-plane magnetization. Our experiments, conducted on multiple devices, reveal a gate-voltage-dependent AHE response, suggesting that the underlying mechanisms responsible for the Hall effect in our system can be tuned via electrostatic gating. A minimal microscopic model constrained by the symmetry of the heterostructure shows that both interfacial spin-orbit coupling and time-reversal symmetry breaking via the exchange interaction from magnetization are responsible for the emergence of the in-plane AHE. Our work highlights the importance of system symmetries and exchange interaction in low-dimensional heterostructures for designing novel and tunable Hall effects in layered quantum systems.

cond-mat.mes-hall↗

arXiv2025

Occupancy World Model for Robots

Understanding and forecasting the scene evolutions deeply affect the exploration and decision of embodied agents. While traditional methods simulate scene evolutions through trajectory prediction of potential instances, current works use the occupancy world model as a generative framework for describing fine-grained overall scene dynamics. However, existing methods cluster on the outdoor structured road scenes, while ignoring the exploration of forecasting 3D occupancy scene evolutions for robots in indoor scenes. In this work, we explore a new framework for learning the scene evolutions of observed fine-grained occupancy and propose an occupancy world model based on the combined spatio-temporal receptive field and guided autoregressive transformer to forecast the scene evolutions, called RoboOccWorld. We propose the Conditional Causal State Attention (CCSA), which utilizes camera poses of next state as conditions to guide the autoregressive transformer to adapt and understand the indoor robotics scenarios. In order to effectively exploit the spatio-temporal cues from historical observations, Hybrid Spatio-Temporal Aggregation (HSTA) is proposed to obtain the combined spatio-temporal receptive field based on multi-scale spatio-temporal windows. In addition, we restructure the OccWorld-ScanNet benchmark based on local annotations to facilitate the evaluation of the indoor 3D occupancy scene evolution prediction task. Experimental results demonstrate that our RoboOccWorld outperforms state-of-the-art methods in indoor 3D occupancy scene evolution prediction task. The code will be released soon.

cs.CV↗

arXiv2025

Testing the Gallium anomaly using Electron-Neutrino Scattering

The Gallium anomaly is an unexplained deficit in the neutrinos observed during the calibration of GALLEX and SAGE using a $^{51}$Cr radioactive source and recently confirmed by BEST. The possible explanations for this deficit include an overestimation of the neutrino absorption cross section in Ga, an incorrect measurement of the source activity or the existence of sterile neutrinos. However, as this deficit has only been observed in Ga detectors, it has not been possible to distinguish among various proposals. Therefore, we propose an experiment using the same radioactive source but with a different detection method, electron-neutrino scattering. We discuss potential locations for such an experiment, estimating the main backgrounds and expected event rates, considering various target masses and source positions. Even if the anomaly does not result from the detection method, such an experiment can provide an independent determination of the branching ratio of the $^{51}$Cr decay by using the spectral information or observing the scattering angle. It is also sensitive to an eventual baseline dependence of the anomaly, as is predicted in sterile neutrino models.

hep-ph↗

arXiv2025

RoboOcc: Enhancing the Geometric and Semantic Scene Understanding for Robots

3D occupancy prediction enables the robots to obtain spatial fine-grained geometry and semantics of the surrounding scene, and has become an essential task for embodied perception. Existing methods based on 3D Gaussians instead of dense voxels do not effectively exploit the geometry and opacity properties of Gaussians, which limits the network's estimation of complex environments and also limits the description of the scene by 3D Gaussians. In this paper, we propose a 3D occupancy prediction method which enhances the geometric and semantic scene understanding for robots, dubbed RoboOcc. It utilizes the Opacity-guided Self-Encoder (OSE) to alleviate the semantic ambiguity of overlapping Gaussians and the Geometry-aware Cross-Encoder (GCE) to accomplish the fine-grained geometric modeling of the surrounding scene. We conduct extensive experiments on Occ-ScanNet and EmbodiedOcc-ScanNet datasets, and our RoboOcc achieves state-of the-art performance in both local and global camera settings. Further, in ablation studies of Gaussian parameters, the proposed RoboOcc outperforms the state-of-the-art methods by a large margin of (8.47, 6.27) in IoU and mIoU metric, respectively. The codes will be released soon.

cs.RO↗

arXiv2025

Jupiter: Fast and Resource-Efficient Collaborative Inference of Generative LLMs on Edge Devices

Generative large language models (LLMs) have garnered significant attention due to their exceptional capabilities in various AI tasks. Traditionally deployed in cloud datacenters, LLMs are now increasingly moving towards more accessible edge platforms to protect sensitive user data and ensure privacy preservation. The limited computational resources of individual edge devices, however, can result in excessively prolonged inference latency and overwhelmed memory usage. While existing research has explored collaborative edge computing to break the resource wall of individual devices, these solutions yet suffer from massive communication overhead and under-utilization of edge resources. Furthermore, they focus exclusively on optimizing the prefill phase, neglecting the crucial autoregressive decoding phase for generative LLMs. To address that, we propose Jupiter, a fast, scalable, and resource-efficient collaborative edge AI system for generative LLM inference. Jupiter introduces a flexible pipelined architecture as a principle and differentiates its system design according to the differentiated characteristics of the prefill and decoding phases. For prefill phase, Jupiter submits a novel intra-sequence pipeline parallelism and develops a meticulous parallelism planning strategy to maximize resource efficiency; For decoding, Jupiter devises an effective outline-based pipeline parallel decoding mechanism combined with speculative decoding, which further magnifies inference acceleration. Extensive evaluation based on realistic implementation demonstrates that Jupiter remarkably outperforms state-of-the-art approaches under various edge environment setups, achieving up to 26.1x end-to-end latency reduction while rendering on-par generation quality.

cs.DC↗

arXiv2025

Fully-inductive Node Classification on Arbitrary Graphs

One fundamental challenge in graph machine learning is generalizing to new graphs. Many existing methods following the inductive setup can generalize to test graphs with new structures, but assuming the feature and label spaces remain the same as the training ones. This paper introduces a fully-inductive setup, where models should perform inference on arbitrary test graphs with new structures, feature and label spaces. We propose GraphAny as the first attempt at this challenging setup. GraphAny models inference on a new graph as an analytical solution to a LinearGNN, which can be naturally applied to graphs with any feature and label spaces. To further build a stronger model with learning capacity, we fuse multiple LinearGNN predictions with learned inductive attention scores. Specifically, the attention module is carefully parameterized as a function of the entropy-normalized distance features between pairs of LinearGNN predictions to ensure generalization to new graphs. Empirically, GraphAny trained on a single Wisconsin dataset with only 120 labeled nodes can generalize to 30 new graphs with an average accuracy of 67.26%, surpassing not only all inductive baselines, but also strong transductive methods trained separately on each of the 30 test graphs.

cs.LG↗

arXiv2025

Development of portable cosmic-ray muon detector array for muography

As the multidisciplinary applications of cosmic-ray muons expand to large-scale and wide-area scenarios, the construction of cosmic-ray muon detector arrays has become a key solution to overcome the hardware limitations of individual detector. For muography, the array-based detector design enables fast-scanning of large target objects, allowing for rapid identification of density variation regions, which can improve the efficiency of tomography. This paper integrates scintillator detector technology with Internet of things (IoT) technology, proposing a novel array networking model for nationwide deployment. The model enables long-distance data collection and distribution, laying the foundation for future multidisciplinary applications such as muography and other fields.

physics.ins-det↗

arXiv2025

Efficient Training of Generalizable Visuomotor Policies via Control-Aware Augmentation

Improving generalization is one key challenge in embodied AI, where obtaining large-scale datasets across diverse scenarios is costly. Traditional weak augmentations, such as cropping and flipping, are insufficient for improving a model's performance in new environments. Existing data augmentation methods often disrupt task-relevant information in images, potentially degrading performance. To overcome these challenges, we introduce EAGLE, an efficient training framework for generalizable visuomotor policies that improves upon existing methods by (1) enhancing generalization by applying augmentation only to control-related regions identified through a self-supervised control-aware mask and (2) improving training stability and efficiency by distilling knowledge from an expert to a visuomotor student policy, which is then deployed to unseen environments without further fine-tuning. Comprehensive experiments on three domains, including the DMControl Generalization Benchmark, the enhanced Robot Manipulation Distraction Benchmark, and a long-sequential drawer-opening task, validate the effectiveness of our method.

cs.RO↗

arXiv2025

Discrete Policy: Learning Disentangled Action Space for Multi-Task Robotic Manipulation

Learning visuomotor policy for multi-task robotic manipulation has been a long-standing challenge for the robotics community. The difficulty lies in the diversity of action space: typically, a goal can be accomplished in multiple ways, resulting in a multimodal action distribution for a single task. The complexity of action distribution escalates as the number of tasks increases. In this work, we propose \textbf{Discrete Policy}, a robot learning method for training universal agents capable of multi-task manipulation skills. Discrete Policy employs vector quantization to map action sequences into a discrete latent space, facilitating the learning of task-specific codes. These codes are then reconstructed into the action space conditioned on observations and language instruction. We evaluate our method on both simulation and multiple real-world embodiments, including both single-arm and bimanual robot settings. We demonstrate that our proposed Discrete Policy outperforms a well-established Diffusion Policy baseline and many state-of-the-art approaches, including ACT, Octo, and OpenVLA. For example, in a real-world multi-task training setting with five tasks, Discrete Policy achieves an average success rate that is 26\% higher than Diffusion Policy and 15\% higher than OpenVLA. As the number of tasks increases to 12, the performance gap between Discrete Policy and Diffusion Policy widens to 32.5\%, further showcasing the advantages of our approach. Our work empirically demonstrates that learning multi-task policies within the latent space is a vital step toward achieving general-purpose agents.

cs.RO↗