arXiv subjects

Ning Liu

Publications and source records attributed to Ning Liu.

At least 91 records · Page 5Linked to original sources

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

An Efficient Private GPT Never Autoregressively Decodes

The wide deployment of the generative pre-trained transformer (GPT) has raised privacy concerns for both clients and servers. While cryptographic primitives can be employed for secure GPT inference to protect the privacy of both parties, they introduce considerable performance overhead.To accelerate secure inference, this study proposes a public decoding and secure verification approach that utilizes public GPT models, motivated by the observation that securely decoding one and multiple tokens takes a similar latency. The client uses the public model to generate a set of tokens, which are then securely verified by the private model for acceptance. The efficiency of our approach depends on the acceptance ratio of tokens proposed by the public model, which we improve from two aspects: (1) a private sampling protocol optimized for cryptographic primitives and (2) model alignment using knowledge distillation. Our approach improves the efficiency of secure decoding while maintaining the same level of privacy and generation quality as standard secure decoding. Experiments demonstrate a $2.1\times \sim 6.0\times$ speedup compared to standard decoding across three pairs of public-private models and different network conditions.

cs.CR↗

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

Nonlinear Evolution Toward the Linear Diffusive Profile in the Presence of Couette Flow

In this paper, we investigate the long-time behavior of solutions to the two-dimensional Navier-Stokes equations with initial data evolving under the influence of the planar Couette flow. We focus on general perturbations, which may be large and of low regularity, including singular configurations such as point vortices, and show that the vorticity asymptotically approaches a constant multiple of the fundamental solution of the corresponding linearized vorticity equation after a long-time evolution determined by the relative Reynolds number.

math.AP↗

arXiv2025

Interplay of $95$ GeV Diphoton Excess and Dark Matter in Supersymmetric Triplet Model

The decay of the Higgs boson and the nature of dark matter remain fundamental challenges in particle physics. We investigate the $95$ GeV diphoton excess and dark matter within the framework of the triplet-extended Minimal Supersymmetric Standard Model (TMSSM). In this model, an additional Hypercharge $Y=0$, $SU(2)_L$ triplet superfield is introduced. Mixing between the triplet and doublet Higgs states enhances the diphoton signal strength of the $95$ GeV Higgs boson, resulting in $μ_{γγ}^{\text{CMS+ATLAS}} = 0.24_{-0.08}^{+0.09}$, which is consistent with experimental observations. This enhancement arises primarily from charged Higgs loop contributions. Additionally, the model accommodates viable dark matter candidates in the form of a bino-dominated neutralino. The relic density is reduced to the observed value through resonance-enhanced annihilation via the Higgs portal or co-annihilation with the triplino or higgsino. This reduction remains consistent with constraints from direct and indirect detection experiments. A comprehensive parameter scan demonstrates that the TMSSM can simultaneously explain the $95$ GeV diphoton excess, the observed $125$ GeV Higgs mass, and the dark matter relic density, establishing a compelling and theoretically consistent framework.

hep-ph↗

arXiv2025

Irreducible characters and bitrace for the $q$-rook monoid

This paper studies irreducible characters of the $q$-rook monoid algebra $R_n(q)$ using the vertex algebraic method. Based on the Frobenius formula for $R_n(q)$, a new iterative character formula is derived with the help of the vertex operator realization of the Schur symmetric function. The same idea also leads to a simple proof of the Murnaghan-Nakayama rule for $R_n(q)$. We also introduce the bitrace for the $q$-rook monoid and derive its combinatorial formula as a generalization of the bitrace formula for the Iwahori-Hecke algebra. The character table of $R_n(q)$ with $|μ|=5$ is listed in the appendix.

math.CO↗

arXiv2025

Probing Gauged $U(1)$ Sub-GeV Dark Matter via Cosmic Ray Cooling in Active Galactic Nuclei

Cosmic rays (CRs) traversing the dark matter (DM) spike surrounding active galactic nuclei (AGNs) can be cooled through interactions with DM particles. In this study, we investigated constraints on sub-GeV DM particles charged under various $U(1)$ gauge symmetries by exploiting the cooling effect of CRs in AGNs. We find that for low DM and mediator masses, the CR cooling rate is higher compared to the standard model cooling process. Furthermore, by utilizing constraints from the CR cooling effect in NGC 1068 and TXS 0506+056, we explore the bounds on the DM-electron and DM-proton elastic scattering cross-sections. Our results indicate that in the sub-GeV DM mass range, these constraints are more stringent than those from certain boosted DM mechanisms and current direct detection limits.

hep-ph↗

arXiv2025

Cooperative Hybrid Multi-Agent Pathfinding Based on Shared Exploration Maps

Multi-Agent Pathfinding is used in areas including multi-robot formations, warehouse logistics, and intelligent vehicles. However, many environments are incomplete or frequently change, making it difficult for standard centralized planning or pure reinforcement learning to maintain both global solution quality and local flexibility. This paper introduces a hybrid framework that integrates D* Lite global search with multi-agent reinforcement learning, using a switching mechanism and a freeze-prevention strategy to handle dynamic conditions and crowded settings. We evaluate the framework in the discrete POGEMA environment and compare it with baseline methods. Experimental outcomes indicate that the proposed framework substantially improves success rate, collision rate, and path efficiency. The model is further tested on the EyeSim platform, where it maintains feasible Pathfinding under frequent changes and large-scale robot deployments.

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↗

arXiv2025

Learning from Imperfect Demonstrations with Self-Supervision for Robotic Manipulation

Improving data utilization, especially for imperfect data from task failures, is crucial for robotic manipulation due to the challenging, time-consuming, and expensive data collection process in the real world. Current imitation learning (IL) typically discards imperfect data, focusing solely on successful expert data. While reinforcement learning (RL) can learn from explorations and failures, the sim2real gap and its reliance on dense reward and online exploration make it difficult to apply effectively in real-world scenarios. In this work, we aim to conquer the challenge of leveraging imperfect data without the need for reward information to improve the model performance for robotic manipulation in an offline manner. Specifically, we introduce a Self-Supervised Data Filtering framework (SSDF) that combines expert and imperfect data to compute quality scores for failed trajectory segments. High-quality segments from the failed data are used to expand the training dataset. Then, the enhanced dataset can be used with any downstream policy learning method for robotic manipulation tasks. Extensive experiments on the ManiSkill2 benchmark built on the high-fidelity Sapien simulator and real-world robotic manipulation tasks using the Franka robot arm demonstrated that the SSDF can accurately expand the training dataset with high-quality imperfect data and improve the success rates for all robotic manipulation tasks.

cs.RO↗

arXiv2025

Phase behaviors and dynamics of active particle systems in double-well potential

In this study, we investigate the behaviors and dynamics of self-propelled particles with active reorientation (AR) in a double-well potential. We explore the competition between AR and external potentials, revealing that self-propelled particles exhibit flocking and clustering behaviors in an asymmetric potential trap. Through molecular dynamics simulations, we obtain a phase diagram that illustrates flocking behavior as a function of active reorientation and potential asymmetry. We compare the responses of inactive and active particles to the potential, finding that active reorientation significantly increases aggregation on one side of the asymmetric potential well. Additionally, by calculating the mean squared displacement and scaling exponent, we identify distinct diffusion regimes. Our findings demonstrate that active particles with active reorientation are more sensitive to variations in double-well potentials.

cond-mat.soft↗

arXiv2025

On $L^p$ extremals for Fourier extension estimate to fractional surface

This article investigates the Fourier extension operator associated with the fractional surface $(ξ,|ξ|^α)$ for $α\geq 2$. We show that the relevant $L^p\to L^q$ Fourier extension inequality possesses extremals for all exponents $p\in[1,2]$. Moreover, for all $p\in(1,2]$, the corresponding $L^p$-extremal sequences are precompact up to symmetries.

math.CA↗

arXiv2025

ChatVLA: Unified Multimodal Understanding and Robot Control with Vision-Language-Action Model

Humans possess a unified cognitive ability to perceive, comprehend, and interact with the physical world. Why can't large language models replicate this holistic understanding? Through a systematic analysis of existing training paradigms in vision-language-action models (VLA), we identify two key challenges: spurious forgetting, where robot training overwrites crucial visual-text alignments, and task interference, where competing control and understanding tasks degrade performance when trained jointly. To overcome these limitations, we propose ChatVLA, a novel framework featuring Phased Alignment Training, which incrementally integrates multimodal data after initial control mastery, and a Mixture-of-Experts architecture to minimize task interference. ChatVLA demonstrates competitive performance on visual question-answering datasets and significantly surpasses state-of-the-art vision-language-action (VLA) methods on multimodal understanding benchmarks. Notably, it achieves a six times higher performance on MMMU and scores 47.2% on MMStar with a more parameter-efficient design than ECoT. Furthermore, ChatVLA demonstrates superior performance on 25 real-world robot manipulation tasks compared to existing VLA methods like OpenVLA. Our findings highlight the potential of our unified framework for achieving both robust multimodal understanding and effective robot control.

cs.RO↗

arXiv2025

SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: human-based, optimization-based, generation-based, and the recent indirect and multilingual jailbreaks. However, delivering a practical jailbreak defense is challenging because it needs to not only handle all the above jailbreak attacks but also incur negligible delays to user prompts, as well as be compatible with both open-source and closed-source LLMs. Inspired by how the traditional security concept of shadow stacks defends against memory overflow attacks, this paper introduces a generic LLM jailbreak defense framework called SelfDefend, which establishes a shadow LLM as a defense instance (in detection state) to concurrently protect the target LLM instance (in normal answering state) in the normal stack and collaborate with it for checkpoint-based access control. The effectiveness of SelfDefend builds upon our observation that existing LLMs can identify harmful prompts or intentions in user queries, which we empirically validate using mainstream GPT-3.5/4 models against major jailbreak attacks. To further improve the defense's robustness and minimize costs, we employ a data distillation approach to tune dedicated open-source defense models. When deployed to protect GPT-3.5/4, Claude, Llama-2-7b/13b, and Mistral, these models outperform seven state-of-the-art defenses and match the performance of GPT-4-based SelfDefend, with significantly lower extra delays. Further experiments show that the tuned models are robust to adaptive jailbreaks and prompt injections.

cs.CR↗

arXiv2025

Stability threshold of the two-dimensional Couette flow in the whole plane

In this paper, we study the stability threshold for the two-dimensional Couette flow in the whole plane. Our main result establishes that the asymptotic stability threshold is at most $\frac{1}{3}+$ for Sobolev perturbations with additional control over low horizontal frequencies, aligning with the threshold results in periodic domains. As a secondary outcome of our approach, we also prove the asymptotic stability for perturbations in weak Sobolev regularity with size $ν^{\frac{1}{2}}$.

math.AP↗

arXiv2025

Personalized Language Model Learning on Text Data Without User Identifiers

In many practical natural language applications, user data are highly sensitive, requiring anonymous uploads of text data from mobile devices to the cloud without user identifiers. However, the absence of user identifiers restricts the ability of cloud-based language models to provide personalized services, which are essential for catering to diverse user needs. The trivial method of replacing an explicit user identifier with a static user embedding as model input still compromises data anonymization. In this work, we propose to let each mobile device maintain a user-specific distribution to dynamically generate user embeddings, thereby breaking the one-to-one mapping between an embedding and a specific user. We further theoretically demonstrate that to prevent the cloud from tracking users via uploaded embeddings, the local distributions of different users should either be derived from a linearly dependent space to avoid identifiability or be close to each other to prevent accurate attribution. Evaluation on both public and industrial datasets using different language models reveals a remarkable improvement in accuracy from incorporating anonymous user embeddings, while preserving real-time inference requirement.

cs.LG↗

arXiv2024

Optical tweezers with optical vortex based on deep learning

Optical tweezers (OTs) with structured light expand degrees of freedom of particle manipulation. However, the studies of structured optical tweezers are usually accompanied by complex theoretical models, strict simulation conditions, and uncertain experimental factors, which may bring about high time costs and insufficiently precise results. In this work, we proposed a bidirectional neural network model for the analysis and design of OTs with optical vortices (OVs) as a typical structured light beam. In analyzing optical forces, the network can achieve over 98% accuracy and improve computational efficiency by more than 20 times. In further analyzing particle trajectories, the network can also achieve over 95.5% accuracy. Meanwhile, in OTs with OV-like beams, our network can still predict particle motion behavior with a high accuracy of up to 96.2%. Our network can inversely design optical vortex tweezers on demand with 95.4% accuracy. In addition, the experimental results in OTs with plasmonic vortex can be analyzed by the proposed model, which can be used to achieve arbitrary optical manipulation. Our work demonstrates that the proposed deep learning network can provide an effective algorithmic platform for the analysis and design of OTs, and is expected to promote the application of OTs in biomedicine.

physics.optics↗

arXiv2024

Scaling Diffusion Policy in Transformer to 1 Billion Parameters for Robotic Manipulation

Diffusion Policy is a powerful technique tool for learning end-to-end visuomotor robot control. It is expected that Diffusion Policy possesses scalability, a key attribute for deep neural networks, typically suggesting that increasing model size would lead to enhanced performance. However, our observations indicate that Diffusion Policy in transformer architecture (\DP) struggles to scale effectively; even minor additions of layers can deteriorate training outcomes. To address this issue, we introduce Scalable Diffusion Transformer Policy for visuomotor learning. Our proposed method, namely \textbf{\methodname}, introduces two modules that improve the training dynamic of Diffusion Policy and allow the network to better handle multimodal action distribution. First, we identify that \DP~suffers from large gradient issues, making the optimization of Diffusion Policy unstable. To resolve this issue, we factorize the feature embedding of observation into multiple affine layers, and integrate it into the transformer blocks. Additionally, our utilize non-causal attention which allows the policy network to \enquote{see} future actions during prediction, helping to reduce compounding errors. We demonstrate that our proposed method successfully scales the Diffusion Policy from 10 million to 1 billion parameters. This new model, named \methodname, can effectively scale up the model size with improved performance and generalization. We benchmark \methodname~across 50 different tasks from MetaWorld and find that our largest \methodname~outperforms \DP~with an average improvement of 21.6\%. Across 7 real-world robot tasks, our ScaleDP demonstrates an average improvement of 36.25\% over DP-T on four single-arm tasks and 75\% on three bimanual tasks. We believe our work paves the way for scaling up models for visuomotor learning. The project page is available at scaling-diffusion-policy.github.io.

cs.RO↗