arXiv subjects

Kai Lu

Publications and source records attributed to Kai Lu.

At least 37 records · Page 2Linked to original sources

arXiv2025

BAGNet: A Boundary-Aware Graph Attention Network for 3D Point Cloud Semantic Segmentation

Since the point cloud data is inherently irregular and unstructured, point cloud semantic segmentation has always been a challenging task. The graph-based method attempts to model the irregular point cloud by representing it as a graph; however, this approach incurs substantial computational cost due to the necessity of constructing a graph for every point within a large-scale point cloud. In this paper, we observe that boundary points possess more intricate spatial structural information and develop a novel graph attention network known as the Boundary-Aware Graph attention Network (BAGNet). On one hand, BAGNet contains a boundary-aware graph attention layer (BAGLayer), which employs edge vertex fusion and attention coefficients to capture features of boundary points, reducing the computation time. On the other hand, BAGNet employs a lightweight attention pooling layer to extract the global feature of the point cloud to maintain model accuracy. Extensive experiments on standard datasets demonstrate that BAGNet outperforms state-of-the-art methods in point cloud semantic segmentation with higher accuracy and less inference time.

cs.CV↗

arXiv2025

MADLLM: Multivariate Anomaly Detection via Pre-trained LLMs

When applying pre-trained large language models (LLMs) to address anomaly detection tasks, the multivariate time series (MTS) modality of anomaly detection does not align with the text modality of LLMs. Existing methods simply transform the MTS data into multiple univariate time series sequences, which can cause many problems. This paper introduces MADLLM, a novel multivariate anomaly detection method via pre-trained LLMs. We design a new triple encoding technique to align the MTS modality with the text modality of LLMs. Specifically, this technique integrates the traditional patch embedding method with two novel embedding approaches: Skip Embedding, which alters the order of patch processing in traditional methods to help LLMs retain knowledge of previous features, and Feature Embedding, which leverages contrastive learning to allow the model to better understand the correlations between different features. Experimental results demonstrate that our method outperforms state-of-the-art methods in various public anomaly detection datasets.

cs.CL↗

arXiv2025

RUNA: Object-level Out-of-Distribution Detection via Regional Uncertainty Alignment of Multimodal Representations

Enabling object detectors to recognize out-of-distribution (OOD) objects is vital for building reliable systems. A primary obstacle stems from the fact that models frequently do not receive supervisory signals from unfamiliar data, leading to overly confident predictions regarding OOD objects. Despite previous progress that estimates OOD uncertainty based on the detection model and in-distribution (ID) samples, we explore using pre-trained vision-language representations for object-level OOD detection. We first discuss the limitations of applying image-level CLIP-based OOD detection methods to object-level scenarios. Building upon these insights, we propose RUNA, a novel framework that leverages a dual encoder architecture to capture rich contextual information and employs a regional uncertainty alignment mechanism to distinguish ID from OOD objects effectively. We introduce a few-shot fine-tuning approach that aligns region-level semantic representations to further improve the model's capability to discriminate between similar objects. Our experiments show that RUNA substantially surpasses state-of-the-art methods in object-level OOD detection, particularly in challenging scenarios with diverse and complex object instances.

cs.CV↗

arXiv2025

DynPoint: Dynamic Neural Point For View Synthesis

The introduction of neural radiance fields has greatly improved the effectiveness of view synthesis for monocular videos. However, existing algorithms face difficulties when dealing with uncontrolled or lengthy scenarios, and require extensive training time specific to each new scenario. To tackle these limitations, we propose DynPoint, an algorithm designed to facilitate the rapid synthesis of novel views for unconstrained monocular videos. Rather than encoding the entirety of the scenario information into a latent representation, DynPoint concentrates on predicting the explicit 3D correspondence between neighboring frames to realize information aggregation. Specifically, this correspondence prediction is achieved through the estimation of consistent depth and scene flow information across frames. Subsequently, the acquired correspondence is utilized to aggregate information from multiple reference frames to a target frame, by constructing hierarchical neural point clouds. The resulting framework enables swift and accurate view synthesis for desired views of target frames. The experimental results obtained demonstrate the considerable acceleration of training time achieved - typically an order of magnitude - by our proposed method while yielding comparable outcomes compared to prior approaches. Furthermore, our method exhibits strong robustness in handling long-duration videos without learning a canonical representation of video content.

cs.CV↗

arXiv2025

Error-mitigated entanglement-assisted quantum process tomography

In the era of noisy intermediate-scale quantum computing, it is of crucial importance to verify quantum processes and extract information. Quantum process tomography is a typical approach, however, both resource-intensive and vulnerable to state preparation and measurement errors. Here, we propose an error-mitigated entanglement-assisted quantum process tomography (EM-EAPT) framework to address these limitations. By leveraging a maximally entangled state to reduce state preparation complexity and integrating error mitigation techniques, our method significantly enhances robustness against SPAM errors. Experimental validation on a superconducting processor demonstrates the efficacy of EM-EAPT for two-qubit and three-qubit quantum processes. Results show more accurate average gate fidelities close to the realistic estimation, achieving 98.1$\pm$ 0.03% for a CNOT gate and 88.1%$\pm$ 0.04% for a cascaded CNOT process after error mitigation, compared to non-mitigated implementations. This work advances practical quantum verification tools for NISQ devices, enabling higher-fidelity characterization of quantum processes under realistic noise conditions.

quant-ph↗

arXiv2025

HyperGo: Probability-based Directed Hybrid Fuzzing

Directed grey-box fuzzing (DGF) is a target-guided fuzzing intended for testing specific targets (e.g., the potential buggy code). Despite numerous techniques proposed to enhance directedness, the existing DGF techniques still face challenges, such as taking into account the difficulty of reaching different basic blocks when designing the fitness metric, and promoting the effectiveness of symbolic execution (SE) when solving the complex constraints in the path to the target. In this paper, we propose a directed hybrid fuzzer called HyperGo. To address the challenges, we introduce the concept of path probability and combine the probability with distance to form an adaptive fitness metric called probability-based distance. By combining the two factors, probability-based distance can adaptively guide DGF toward paths that are closer to the target and have more easy-to-satisfy path constraints. Then, we put forward an Optimized Symbolic Execution Complementary (OSEC) scheme to combine DGF and SE in a complementary manner. The OSEC would prune the unreachable branches and unsolvable branches, and prioritize symbolic execution of the seeds whose paths are closer to the target and have more branches that are difficult to be covered by DGF. We evaluated HyperGo on 2 benchmarks consisting of 21 programs with a total of 100 target sites. The experimental results show that HyperGo achieves 38.47$\times$, 30.89$\times$, 28.52$\times$, 106.09$\times$ and 143.22$\times$ speedup compared to AFLGo, AFLGoSy, BEACON, WindRanger, and ParmeSan, respectively in reaching target sites, and 3.44$\times$, 3.63$\times$, 4.10$\times$, 3.26$\times$, and 3.00$\times$ speedup in exposing known vulnerabilities. Moreover, HyperGo discovered 37 undisclosed vulnerabilities from 7 real-world programs.

cs.CR↗

arXiv2025

Telegraph flux noise induced beating Ramsey fringe in transmon qubits

Ramsey oscillations typically exhibit an exponential decay envelope due to environmental noise. However, recent experiments have observed nonmonotonic Ramsey fringes characterized by beating patterns, which deviate from the standard behavior. These beating patterns have primarily been attributed to charge-noise fluctuations. In this paper, we investigate the flux-noise origin of these nonmonotonic Ramsey fringes in frequency-tunable transmon qubits. We develop a random telegraph noise (RTN) model to simulate the impact of telegraph-like flux-noise sources on Ramsey oscillations. Our simulations demonstrate that strong flux-RTN sources can induce beating patterns in the Ramsey fringes, showing excellent agreement with experimental observations in transmon qubits influenced by electronic environment-induced flux-noise. Our findings provide valuable insights into the role of flux-noise in qubit decoherence and underscore the importance of considering flux-noise RTN when analyzing nonmonotonic Ramsey fringes.

quant-ph↗

arXiv2024

RefSAM3D: Adapting SAM with Cross-modal Reference for 3D Medical Image Segmentation

The Segment Anything Model (SAM), originally built on a 2D Vision Transformer (ViT), excels at capturing global patterns in 2D natural images but struggles with 3D medical imaging modalities like CT and MRI. These modalities require capturing spatial information in volumetric space for tasks such as organ segmentation and tumor quantification. To address this challenge, we introduce RefSAM3D, which adapts SAM for 3D medical imaging by incorporating a 3D image adapter and cross-modal reference prompt generation. Our approach modifies the visual encoder to handle 3D inputs and enhances the mask decoder for direct 3D mask generation. We also integrate textual prompts to improve segmentation accuracy and consistency in complex anatomical scenarios. By employing a hierarchical attention mechanism, our model effectively captures and integrates information across different scales. Extensive evaluations on multiple medical imaging datasets demonstrate the superior performance of RefSAM3D over state-of-the-art methods. Our contributions advance the application of SAM in accurately segmenting complex anatomical structures in medical imaging.

cs.CV↗

arXiv2024

Hammer: Towards Efficient Hot-Cold Data Identification via Online Learning

Efficient management of storage resources in big data and cloud computing environments requires accurate identification of data's "cold" and "hot" states. Traditional methods, such as rule-based algorithms and early AI techniques, often struggle with dynamic workloads, leading to low accuracy, poor adaptability, and high operational overhead. To address these issues, we propose a novel solution based on online learning strategies. Our approach dynamically adapts to changing data access patterns, achieving higher accuracy and lower operational costs. Rigorous testing with both synthetic and real-world datasets demonstrates a significant improvement, achieving a 90% accuracy rate in hot-cold classification. Additionally, the computational and storage overheads are considerably reduced.

cs.LG↗

arXiv2024

SpatialPIN: Enhancing Spatial Reasoning Capabilities of Vision-Language Models through Prompting and Interacting 3D Priors

Current state-of-the-art spatial reasoning-enhanced VLMs are trained to excel at spatial visual question answering (VQA). However, we believe that higher-level 3D-aware tasks, such as articulating dynamic scene changes and motion planning, require a fundamental and explicit 3D understanding beyond current spatial VQA datasets. In this work, we present SpatialPIN, a framework designed to enhance the spatial reasoning capabilities of VLMs through prompting and interacting with priors from multiple 3D foundation models in a zero-shot, training-free manner. Extensive experiments demonstrate that our spatial reasoning-imbued VLM performs well on various forms of spatial VQA and can extend to help in various downstream robotics tasks such as pick and stack and trajectory planning.

cs.CV↗

arXiv2024

INSTILLER: Towards Efficient and Realistic RTL Fuzzing

Bugs exist in hardware, such as CPU. Unlike software bugs, these hardware bugs need to be detected before deployment. Previous fuzzing work in CPU bug detection has several disadvantages, e.g., the length of RTL input instructions keeps growing, and longer inputs are ineffective for fuzzing. In this paper, we propose INSTILLER (Instruction Distiller), an RTL fuzzer based on ant colony optimization (ACO). First, to keep the input instruction length short and efficient in fuzzing, it distills input instructions with a variant of ACO (VACO). Next, related work cannot simulate realistic interruptions well in fuzzing, and INSTILLER solves the problem of inserting interruptions and exceptions in generating the inputs. Third, to further improve the fuzzing performance of INSTILLER, we propose hardware-based seed selection and mutation strategies. We implement a prototype and conduct extensive experiments against state-of-the-art fuzzing work in real-world target CPU cores. In experiments, INSTILLER has 29.4% more coverage than DiFuzzRTL. In addition, 17.0% more mismatches are detected by INSTILLER. With the VACO algorithm, INSTILLER generates 79.3% shorter input instructions than DiFuzzRTL, demonstrating its effectiveness in distilling the input instructions. In addition, the distillation leads to a 6.7% increase in execution speed on average.

cs.CR↗

arXiv2024

MobFuzz: Adaptive Multi-objective Optimization in Gray-box Fuzzing

Coverage-guided gray-box fuzzing (CGF) is an efficient software testing technique. There are usually multiple objectives to optimize in CGF. However, existing CGF methods cannot successfully find the optimal values for multiple objectives simultaneously. In this paper, we propose a gray-box fuzzer for multi-objective optimization (MOO) called MobFuzz. We model the multi-objective optimization process as a multi-player multi-armed bandit (MPMAB). First, it adaptively selects the objective combination that contains the most appropriate objectives for the current situation. Second, our model deals with the power schedule, which adaptively allocates energy to the seeds under the chosen objective combination. In MobFuzz, we propose an evolutionary algorithm called NIC to optimize our chosen objectives simultaneously without incurring additional performance overhead. To prove the effectiveness of MobFuzz, we conduct experiments on 12 real-world programs and the MAGMA data set. Experiment results show that multi-objective optimization in MobFuzz outperforms single-objective fuzzing in the baseline fuzzers. In contrast to them, MobFuzz can select the optimal objective combination and increase the values of multiple objectives up to 107%, with at most a 55% reduction in the energy consumption. Moreover, MobFuzz has up to 6% more program coverage and finds 3x more unique bugs than the baseline fuzzers. The NIC algorithm has at least a 2x improvement with a performance overhead of approximately 3%.

cs.CR↗

arXiv2024

Value-Driven Mixed-Precision Quantization for Patch-Based Inference on Microcontrollers

Deploying neural networks on microcontroller units (MCUs) presents substantial challenges due to their constrained computation and memory resources. Previous researches have explored patch-based inference as a strategy to conserve memory without sacrificing model accuracy. However, this technique suffers from severe redundant computation overhead, leading to a substantial increase in execution latency. A feasible solution to address this issue is mixed-precision quantization, but it faces the challenges of accuracy degradation and a time-consuming search time. In this paper, we propose QuantMCU, a novel patch-based inference method that utilizes value-driven mixed-precision quantization to reduce redundant computation. We first utilize value-driven patch classification (VDPC) to maintain the model accuracy. VDPC classifies patches into two classes based on whether they contain outlier values. For patches containing outlier values, we apply 8-bit quantization to the feature maps on the dataflow branches that follow. In addition, for patches without outlier values, we utilize value-driven quantization search (VDQS) on the feature maps of their following dataflow branches to reduce search time. Specifically, VDQS introduces a novel quantization search metric that takes into account both computation and accuracy, and it employs entropy as an accuracy representation to avoid additional training. VDQS also adopts an iterative approach to determine the bitwidth of each feature map to further accelerate the search process. Experimental results on real-world MCU devices show that QuantMCU can reduce computation by 2.2x on average while maintaining comparable model accuracy compared to the state-of-the-art patch-based inference methods.

cs.CV↗

arXiv2023

The Progress, Challenges, and Perspectives of Directed Greybox Fuzzing

Greybox fuzzing is a scalable and practical approach for software testing. Most greybox fuzzing tools are coverage-guided as reaching high code coverage is more likely to find bugs. However, since most covered codes may not contain bugs, blindly extending code coverage is less efficient, especially for corner cases. Unlike coverage-guided greybox fuzzing which increases code coverage in an undirected manner, directed greybox fuzzing (DGF) spends most of its time allocation on reaching specific targets (e.g., the bug-prone zone) without wasting resources stressing unrelated parts. Thus, DGF is particularly suitable for scenarios such as patch testing,bug reproduction, and special bug detection. For now, DGF has become an active research area. However, DGF has general limitations and challenges that are worth further studying. Based on the investigation of 42 state-of-the-art fuzzers that are closely related to DGF, we conduct the first in-depth study to summarize the empirical evidence on the research progress of DGF. This paper studies DGF from a broader view, which takes into account not only the location-directed type that targets specific code parts, but also the behavior-directed type that aims to expose abnormal program behaviors. By analyzing the benefits and limitations of DGF research, we try to identify gaps in current research, meanwhile, reveal new research opportunities, and suggest areas for further investigation.

cs.CR↗

arXiv2023

Multi-body SE(3) Equivariance for Unsupervised Rigid Segmentation and Motion Estimation

A truly generalizable approach to rigid segmentation and motion estimation is fundamental to 3D understanding of articulated objects and moving scenes. In view of the closely intertwined relationship between segmentation and motion estimates, we present an SE(3) equivariant architecture and a training strategy to tackle this task in an unsupervised manner. Our architecture is composed of two interconnected, lightweight heads. These heads predict segmentation masks using point-level invariant features and estimate motion from SE(3) equivariant features, all without the need for category information. Our training strategy is unified and can be implemented online, which jointly optimizes the predicted segmentation and motion by leveraging the interrelationships among scene flow, segmentation mask, and rigid transformations. We conduct experiments on four datasets to demonstrate the superiority of our method. The results show that our method excels in both model performance and computational efficiency, with only 0.25M parameters and 0.92G FLOPs. To the best of our knowledge, this is the first work designed for category-agnostic part-level SE(3) equivariance in dynamic point clouds.

cs.CV↗

arXiv2023

Shoggoth: Towards Efficient Edge-Cloud Collaborative Real-Time Video Inference via Adaptive Online Learning

This paper proposes Shoggoth, an efficient edge-cloud collaborative architecture, for boosting inference performance on real-time video of changing scenes. Shoggoth uses online knowledge distillation to improve the accuracy of models suffering from data drift and offloads the labeling process to the cloud, alleviating constrained resources of edge devices. At the edge, we design adaptive training using small batches to adapt models under limited computing power, and adaptive sampling of training frames for robustness and reducing bandwidth. The evaluations on the realistic dataset show 15%-20% model accuracy improvement compared to the edge-only strategy and fewer network costs than the cloud-only strategy.

cs.CV↗

arXiv2023

PAI at SemEval-2023 Task 2: A Universal System for Named Entity Recognition with External Entity Information

The MultiCoNER II task aims to detect complex, ambiguous, and fine-grained named entities in low-context situations and noisy scenarios like the presence of spelling mistakes and typos for multiple languages. The task poses significant challenges due to the scarcity of contextual information, the high granularity of the entities(up to 33 classes), and the interference of noisy data. To address these issues, our team {\bf PAI} proposes a universal Named Entity Recognition (NER) system that integrates external entity information to improve performance. Specifically, our system retrieves entities with properties from the knowledge base (i.e. Wikipedia) for a given text, then concatenates entity information with the input sentence and feeds it into Transformer-based models. Finally, our system wins 2 first places, 4 second places, and 1 third place out of 13 tracks. The code is publicly available at \url{https://github.com/diqiuzhuanzhuan/semeval-2023}.

cs.CL↗

arXiv2023

Decoupling Skill Learning from Robotic Control for Generalizable Object Manipulation

Recent works in robotic manipulation through reinforcement learning (RL) or imitation learning (IL) have shown potential for tackling a range of tasks e.g., opening a drawer or a cupboard. However, these techniques generalize poorly to unseen objects. We conjecture that this is due to the high-dimensional action space for joint control. In this paper, we take an alternative approach and separate the task of learning 'what to do' from 'how to do it' i.e., whole-body control. We pose the RL problem as one of determining the skill dynamics for a disembodied virtual manipulator interacting with articulated objects. The whole-body robotic kinematic control is optimized to execute the high-dimensional joint motion to reach the goals in the workspace. It does so by solving a quadratic programming (QP) model with robotic singularity and kinematic constraints. Our experiments on manipulating complex articulated objects show that the proposed approach is more generalizable to unseen objects with large intra-class variations, outperforming previous approaches. The evaluation results indicate that our approach generates more compliant robotic motion and outperforms the pure RL and IL baselines in task success rates. Additional information and videos are available at https://kl-research.github.io/decoupskill

cs.RO↗