arXiv subjects

Ziming Wang

Publications and source records attributed to Ziming Wang.

At least 91 records · Page 5Linked to original sources

arXiv2025

FAST-LIVO2 on Resource-Constrained Platforms: LiDAR-Inertial-Visual Odometry with Efficient Memory and Computation

This paper presents a lightweight LiDAR-inertial-visual odometry system optimized for resource-constrained platforms. It integrates a degeneration-aware adaptive visual frame selector into error-state iterated Kalman filter (ESIKF) with sequential updates, improving computation efficiency significantly while maintaining a similar level of robustness. Additionally, a memory-efficient mapping structure combining a locally unified visual-LiDAR map and a long-term visual map achieves a good trade-off between performance and memory usage. Extensive experiments on x86 and ARM platforms demonstrate the system's robustness and efficiency. On the Hilti dataset, our system achieves a 33% reduction in per-frame runtime and 47% lower memory usage compared to FAST-LIVO2, with only a 3 cm increase in RMSE. Despite this slight accuracy trade-off, our system remains competitive, outperforming state-of-the-art (SOTA) LIO methods such as FAST-LIO2 and most existing LIVO systems. These results validate the system's capability for scalable deployment on resource-constrained edge computing platforms.

cs.RO↗

arXiv2025

Procedural Fairness and Its Relationship with Distributive Fairness in Machine Learning

Fairness in machine learning (ML) has garnered significant attention in recent years. While existing research has predominantly focused on the distributive fairness of ML models, there has been limited exploration of procedural fairness. This paper proposes a novel method to achieve procedural fairness during the model training phase. The effectiveness of the proposed method is validated through experiments conducted on one synthetic and six real-world datasets. Additionally, this work studies the relationship between procedural fairness and distributive fairness in ML models. On one hand, the impact of dataset bias and the procedural fairness of ML model on its distributive fairness is examined. The results highlight a significant influence of both dataset bias and procedural fairness on distributive fairness. On the other hand, the distinctions between optimizing procedural and distributive fairness metrics are analyzed. Experimental results demonstrate that optimizing procedural fairness metrics mitigates biases introduced or amplified by the decision-making process, thereby ensuring fairness in the decision-making process itself, as well as improving distributive fairness. In contrast, optimizing distributive fairness metrics encourages the ML model's decision-making process to favor disadvantaged groups, counterbalancing the inherent preferences for advantaged groups present in the dataset and ultimately achieving distributive fairness.

cs.LG↗

arXiv2024

Event-Triggered Observer-Based Fixed-Time Consensus Control for Uncertain Nonlinear Multiagent Systems with Unknown States

This paper introduces a novel approach for achieving fixed-time tracking consensus control in multiagent systems (MASs). Departing from the reliance on traditional controllers, our innovative controller integrates modified tuning and Lyapunov functions to guarantee stability and convergence. Furthermore, we have implemented an event-triggered strategy aimed at reducing the frequency of updates, alongside an output-feedback observer to manage unmeasured states effectively. To address the challenges posed by unknown functions and algebraic-loop problems, we opted for radial basis function neural networks (RBF NNs), chosen for their superior performance. Our methodology successfully mitigates Zeno's behavior and ensures stability within a narrowly defined set. The efficacy of our proposed solution is validated through two illustrative simulation examples.

eess.SY↗

arXiv2024

Rethinking Structure Learning For Graph Neural Networks

To improve the performance of Graph Neural Networks (GNNs), Graph Structure Learning (GSL) has been extensively applied to reconstruct or refine original graph structures, effectively addressing issues like heterophily, over-squashing, and noisy structures. While GSL is generally thought to improve GNN performance, it often leads to longer training times and more hyperparameter tuning. Besides, the distinctions among current GSL methods remain ambiguous from the perspective of GNN training, and there is a lack of theoretical analysis to quantify their effectiveness. Recent studies further suggest that, under fair comparisons with the same hyperparameter tuning, GSL does not consistently outperform baseline GNNs. This motivates us to ask a critical question: is GSL really useful for GNNs? To address this question, this paper makes two key contributions. First, we propose a new GSL framework, which includes three steps: GSL base (the representation used for GSL) construction, new structure construction, and view fusion, to better understand the effectiveness of GSL in GNNs. Second, after graph convolution, we analyze the differences in mutual information (MI) between node representations derived from the original topology and those from the newly constructed topology. Surprisingly, our empirical observations and theoretical analysis show that no matter which type of graph structure construction methods are used, after feeding the same GSL bases to the newly constructed graph, there is no MI gain compared to the original GSL bases. To fairly reassess the effectiveness of GSL, we conduct ablation experiments and find that it is the pretrained GSL bases that enhance GNN performance, and in most cases, GSL cannot improve GNN performance. This finding encourages us to rethink the essential components in GNNs, such as self-training and structural encoding, in GNN design rather than GSL.

cs.LG↗

arXiv2024

SE(3)-bi-equivariant Transformers for Point Cloud Assembly

Given a pair of point clouds, the goal of assembly is to recover a rigid transformation that aligns one point cloud to the other. This task is challenging because the point clouds may be non-overlapped, and they may have arbitrary initial positions. To address these difficulties, we propose a method, called SE(3)-bi-equivariant transformer (BITR), based on the SE(3)-bi-equivariance prior of the task: it guarantees that when the inputs are rigidly perturbed, the output will transform accordingly. Due to its equivariance property, BITR can not only handle non-overlapped PCs, but also guarantee robustness against initial positions. Specifically, BITR first extracts features of the inputs using a novel $SE(3) \times SE(3)$-transformer, and then projects the learned feature to group SE(3) as the output. Moreover, we theoretically show that swap and scale equivariances can be incorporated into BITR, thus it further guarantees stable performance under scaling and swapping the inputs. We experimentally show the effectiveness of BITR in practical tasks.

cs.AI↗

arXiv2024

Vetting quark-star models with gravitational waves in the hierarchical Bayesian framework

The recent discovery of gravitational waves (GWs) has opened a new avenue for investigating the equation of state (EOS) of dense matter in compact stars, which is an outstanding problem in astronomy and nuclear physics. In the future, next-generation (XG) GW detectors will be constructed, deemed to provide a large number of high-precision observations. We investigate the potential of constraining the EOS of quark stars (QSs) with high-precision measurements of mass $m$ and tidal deformability $Λ$ from the XG GW observatories. We adopt the widely-used bag model for QSs, consisting of four microscopic parameters: the effective bag constant $B_{\rm eff}$, the perturbative quantum chromodynamics correction parameter $a_4$, the strange quark mass $m_s$, and the pairing energy gap $Δ$. With the help of hierarchical Bayesian inference, for the first time we are able to infer the EOS of QSs combining multiple GW observations. Using the top 25 loudest GW events in our simulation, we find that, the constraints on $B_{\rm eff}$ and $Δ$ are tightened by several times, while $a_4$ and $m_s$ are still poorly constrained. We also study a simplified 2-dimensional (2-d) EOS model which was recently proposed in literature. The 2-d model is found to exhibit significant parameter-estimation biases as more GW events are analyzed, while the predicted $m$-$Λ$ relation remains consistent with the full model.

astro-ph.HE↗

arXiv2024

Disentangling Heterogeneous Knowledge Concept Embedding for Cognitive Diagnosis on Untested Knowledge

Cognitive diagnosis is a fundamental and critical task in learning assessment, which aims to infer students' proficiency on knowledge concepts from their response logs. Current works assume each knowledge concept will certainly be tested and covered by multiple exercises. However, whether online or offline courses, it's hardly feasible to completely cover all knowledge concepts in several exercises. Restricted tests lead to undiscovered knowledge deficits, especially untested knowledge concepts(UKCs). In this paper, we propose a novel framework for Cognitive Diagnosis called Disentangling Heterogeneous Knowledge Cognitive Diagnosis(DisKCD) on untested knowledge. Specifically, we leverage course grades, exercise questions, and learning resources to learn the potential representations of students, exercises, and knowledge concepts. In particular, knowledge concepts are disentangled into tested and untested based on the limiting actual exercises. We construct a heterogeneous relation graph network via students, exercises, tested knowledge concepts(TKCs), and UKCs. Then, through a hierarchical heterogeneous message-passing mechanism, the fine-grained relations are incorporated into the embeddings of the entities. Finally, the embeddings will be applied to multiple existing cognitive diagnosis models to infer students' proficiency on UKCs. Experimental results on real-world datasets show that the proposed model can effectively improve the performance of the task of diagnosing students' proficiency on UKCs. Our code is available at https://github.com/Hubuers/DisKCD.

cs.AI↗

arXiv2024

Perspective: Floquet engineering topological states from effective models towards realistic materials

With significant advances in classifying and cataloguing topological matter, the focus of topological physics has shifted towards quantum control, particularly the creation and manipulation of topological phases of matter. Floquet engineering, the concept of tailoring a system by periodic fields, offers a powerful tool to manipulate electronic properties of condensed systems, and even to create exotic non-equilibrium topological states that are impossibly present in equilibrium scenarios. In this perspective, we give a brief review of recent progress in theoretical investigations of Floquet engineering topological states from effective models towards realistic materials. We show that light irradiation can realize various desired topological states through the introduction of symmetry breaking, such as first- and higher-order Weyl fermions, quadrupole topological insulator with periodic driving and disorder, quantum anomalous Hall effects with a tunable Chern number, as well as beyond. Moreover, based on first-principles calculations and Floquet theorem, we show several realistic material candidates proposed as potential hosts for promising Floquet topological states, facilitating their verification in experiments. We believe that our perspective on Floquet engineering of topological states will advance further studies of rich exotic light-induced phenomena in condensed matter physics.

physics.comp-ph↗

arXiv2024

EAS-SNN: End-to-End Adaptive Sampling and Representation for Event-based Detection with Recurrent Spiking Neural Networks

Event cameras, with their high dynamic range and temporal resolution, are ideally suited for object detection, especially under scenarios with motion blur and challenging lighting conditions. However, while most existing approaches prioritize optimizing spatiotemporal representations with advanced detection backbones and early aggregation functions, the crucial issue of adaptive event sampling remains largely unaddressed. Spiking Neural Networks (SNNs), which operate on an event-driven paradigm through sparse spike communication, emerge as a natural fit for addressing this challenge. In this study, we discover that the neural dynamics of spiking neurons align closely with the behavior of an ideal temporal event sampler. Motivated by this insight, we propose a novel adaptive sampling module that leverages recurrent convolutional SNNs enhanced with temporal memory, facilitating a fully end-to-end learnable framework for event-based detection. Additionally, we introduce Residual Potential Dropout (RPD) and Spike-Aware Training (SAT) to regulate potential distribution and address performance degradation encountered in spike-based sampling modules. Empirical evaluation on neuromorphic detection datasets demonstrates that our approach outperforms existing state-of-the-art spike-based methods with significantly fewer parameters and time steps. For instance, our method yields a 4.4\% mAP improvement on the Gen1 dataset, while requiring 38\% fewer parameters and only three time steps. Moreover, the applicability and effectiveness of our adaptive sampling methodology extend beyond SNNs, as demonstrated through further validation on conventional non-spiking models. Code is available at https://github.com/Windere/EAS-SNN.

cs.CV↗

arXiv2024

Event-Triggered Optimal Tracking Control for Strict-Feedback Nonlinear Systems With Non-Affine Nonlinear Faults

This article studies the control ideas of the optimal backstepping technique, proposing an event-triggered optimal tracking control scheme for a class of strict-feedback nonlinear systems with non-affine and nonlinear faults. A simplified identifier-critic-actor framework is employed in the reinforcement learning algorithm to achieve optimal control. The identifier estimates the unknown dynamic functions, the critic evaluates the system performance, and the actor implements control actions, enabling modeling and control of anonymous systems for achieving optimal control performance. In this paper, a simplified reinforcement learning algorithm is designed by deriving update rules from the negative gradient of a simple positive function related to the Hamilton-Jacobi-Bellman equation, and it also releases the stringent persistent excitation condition. Then, a fault-tolerant control method is developed by applying filtered signals for controller design. Additionally, to address communication resource reduction, an event-triggered mechanism is employed for designing the actual controller. Finally, the proposed scheme's feasibility is validated through theoretical analysis and simulation.

math.OC↗

arXiv2024

GWnext 2024: Meeting Summary

GWnext 2024 was a meeting held in the Kavli Institute for Astronomy and Astrophysics at Peking University in March $4^\text{th} - 8^\text{th}$, 2024. In the meeting researchers at different career stages -- with a particular focus on early career scientists -- working on the different aspects of gravitational wave (GW) astronomy gathered to discuss the current status as well as prospects of the field. The meeting was divided into three core sessions: Astrophysics, GW Theory, and Detection. Each session consisted of introductory talks and extended discussion sessions. Moreover, there was a poster session where students could present their results. In this paper, we summarize the results presented during the meeting and present the most important outcomes.

astro-ph.HE↗

arXiv2024

ConvLLaVA: Hierarchical Backbones as Visual Encoder for Large Multimodal Models

High-resolution Large Multimodal Models (LMMs) encounter the challenges of excessive visual tokens and quadratic visual complexity. Current high-resolution LMMs address the quadratic complexity while still generating excessive visual tokens. However, the redundancy in visual tokens is the key problem as it leads to more substantial compute. To mitigate this issue, we propose ConvLLaVA, which employs ConvNeXt, a hierarchical backbone, as the visual encoder of LMM to replace Vision Transformer (ViT). ConvLLaVA compresses high-resolution images into information-rich visual features, effectively preventing the generation of excessive visual tokens. To enhance the capabilities of ConvLLaVA, we propose two critical optimizations. Since the low-resolution pretrained ConvNeXt underperforms when directly applied on high resolution, we update it to bridge the gap. Moreover, since ConvNeXt's original compression ratio is inadequate for much higher resolution inputs, we train a successive stage to further compress the visual tokens, thereby reducing redundancy. These optimizations enable ConvLLaVA to support inputs of 1536x1536 resolution generating only 576 visual tokens, capable of handling images of arbitrary aspect ratios. Experimental results demonstrate that our method achieves competitive performance with state-of-the-art models on mainstream benchmarks. The ConvLLaVA model series are publicly available at https://github.com/alibaba/conv-llava.

cs.CV↗

arXiv2024

Hybrid Event-triggered Control of Nonlinear System with Full State Constraints and Disturbance

This article focuses on the problem of adaptive tracking control for a specific type of nonlinear system that is subject to full-state constraints via a hybrid event-triggered control (HETC) strategy. With the auxiliary system, we proposed a 'log' function to deal with the full-state constraint. Additionally, a disturbance observer (DO) is constructed to handle the unmeasurable external disturbance. Then, by employing radial basis function neural networks (RBFNNs) and a first-order differentiator, an opportune backstepping design procedure is given to avoid the problem of "explosion of complexity". The HETC strategy, including the fixed and relative threshold, is presented to provide more flexibility in balancing the system performances and network burdens. Finally, to demonstrate the effectiveness of the aforementioned control scheme, a simulation example is presented to validate its effectiveness.

eess.SY↗

arXiv2024

The Impact of Spin in Compact Binary Foreground Subtraction for Estimating the Residual Stochastic Gravitational-wave Background in Ground-based Detectors

Stochastic gravitational-wave (GW) background (SGWB) contains information about the early Universe and astrophysical processes. The recent evidence of SGWB by pulsar timing arrays in the nanohertz band is a breakthrough in the GW astronomy. For ground-based GW detectors, while in data analysis, the SGWB can be masked by loud GW events from compact binary coalescences (CBCs). Assuming a next-generation ground-based GW detector network, we investigate the potential for detecting the astrophysical and cosmological SGWB with non-CBC origins by subtracting recovered foreground signals of loud CBC events. The Fisher Information Matrix (FIM) method is adopted for quick calculation. As an extension of the studies by Sachdev {\it et al.} (2020) and Zhou {\it et al.} (2023), two more essential features are considered. Firstly, we incorporate non-zero aligned or anti-aligned spin parameters in our waveform model. Because of the inclusion of spins, we obtain significantly more pessimistic results than the previous work, where the residual energy density of foreground is even larger than the original CBC foreground. For the most extreme case, we observe that the subtraction results are approximately 10 times worse for binary black hole events and 20 times worse for binary neutron star events than the scenarios without accounting for spins. The degeneracy between the spin parameters and the symmetric mass ratio is strong in the parameter estimation process, and it contributes most to the imperfect foreground subtraction. Secondly, in this work, extreme CBC events with condition numbers of FIMs $c_{\rmΓ}>10^{15}$ are preserved. The impacts of these extreme events on foreground subtraction are discussed. Our results have important implications for assessing the detectability of SGWB from non-CBC origins for ground-based GW detectors.

gr-qc↗

arXiv2024

Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRL

Large Language Models (LLMs) play a crucial role in capturing structured semantics to enhance language understanding, improve interpretability, and reduce bias. Nevertheless, an ongoing controversy exists over the extent to which LLMs can grasp structured semantics. To assess this, we propose using Semantic Role Labeling (SRL) as a fundamental task to explore LLMs' ability to extract structured semantics. In our assessment, we employ the prompting approach, which leads to the creation of our few-shot SRL parser, called PromptSRL. PromptSRL enables LLMs to map natural languages to explicit semantic structures, which provides an interpretable window into the properties of LLMs. We find interesting potential: LLMs can indeed capture semantic structures, and scaling-up doesn't always mirror potential. Additionally, limitations of LLMs are observed in C-arguments, etc. Lastly, we are surprised to discover that significant overlap in the errors is made by both LLMs and untrained humans, accounting for almost 30% of all errors.

cs.CL↗

arXiv2024

GRSN: Gated Recurrent Spiking Neurons for POMDPs and MARL

Spiking neural networks (SNNs) are widely applied in various fields due to their energy-efficient and fast-inference capabilities. Applying SNNs to reinforcement learning (RL) can significantly reduce the computational resource requirements for agents and improve the algorithm's performance under resource-constrained conditions. However, in current spiking reinforcement learning (SRL) algorithms, the simulation results of multiple time steps can only correspond to a single-step decision in RL. This is quite different from the real temporal dynamics in the brain and also fails to fully exploit the capacity of SNNs to process temporal data. In order to address this temporal mismatch issue and further take advantage of the inherent temporal dynamics of spiking neurons, we propose a novel temporal alignment paradigm (TAP) that leverages the single-step update of spiking neurons to accumulate historical state information in RL and introduces gated units to enhance the memory capacity of spiking neurons. Experimental results show that our method can solve partially observable Markov decision processes (POMDPs) and multi-agent cooperation problems with similar performance as recurrent neural networks (RNNs) but with about 50% power consumption.

cs.NE↗

arXiv2024

Qilin-Med: Multi-stage Knowledge Injection Advanced Medical Large Language Model

Integrating large language models (LLMs) into healthcare holds great potential but faces challenges. Pre-training LLMs from scratch for domains like medicine is resource-heavy and often unfeasible. On the other hand, sole reliance on Supervised Fine-tuning (SFT) can result in overconfident predictions and may not tap into domain-specific insights. In response, we present a multi-stage training method combining Domain-specific Continued Pre-training (DCPT), SFT, and Direct Preference Optimization (DPO). In addition, we publish a 3Gb Chinese Medicine (ChiMed) dataset, encompassing medical question answering, plain texts, knowledge graphs, and dialogues, segmented into three training stages. The medical LLM trained with our pipeline, Qilin-Med, shows substantial performance improvement. In the CPT and SFT phases, Qilin-Med achieved 38.4% and 40.0% accuracy on the CMExam test set, respectively. It outperformed the basemodel Baichuan-7B (accuracy: 33.5%), by 7.5%. In the DPO phase, it scored 16.66 in BLEU-1 and 27.44 in ROUGE-1 on the Huatuo-26M test set, bringing further improvement to the SFT phase (12.69 in BLEU-1 and 24.21 in ROUGE-1). Additionally, we have further enhanced the model's performance through the Retrieval Augmented Generation (RAG) approach. Experiments demonstrate that Qilin-Med-RAG achieves an accuracy rate of 42.8% on CMExam. These results highlight the contribution of our novel training approach in building LLMs for medical applications.

cs.CL↗

arXiv2024

An Efficient Sparse Identification Algorithm For Stochastic Systems With General Observation Sequences

This paper studies the sparse identification problem of unknown sparse parameter vectors in stochastic dynamic systems. Firstly, a novel sparse identification algorithm is proposed, which can generate sparse estimates based on least squares estimation by adaptively adjusting the threshold. Secondly, under a possibly weakest non-persistent excited condition, we prove that the proposed algorithm can correctly identify the zero and nonzero elements of the sparse parameter vector using a finite number of observations, and further estimates of the nonzero elements almost surely converge to the true values. Compared with the related works, e.g., LASSO, our method only requires the weakest assumptions and does not require solving additional optimization problems. Besides, our theoretical results do not require any statistical assumptions on the regression signals, including independence or stationarity, which makes our results promising for application to stochastic feedback systems. Thirdly, the number of finite observations that guarantee the convergence of the zero-element set of unknown sparse parameters of the Hammerstein system is derived for the first time. Finally, numerical simulations are provided, demonstrating the effectiveness of the proposed method. Since there is no additional optimization problem, i.e., no additional numerical error, the proposed algorithm performs much better than other related algorithms.

math.OC↗