arXiv subjects

Hao Fu

Publications and source records attributed to Hao Fu.

At least 55 records · Page 3Linked to original sources

arXiv2025

6D Pose Estimation on Point Cloud Data through Prior Knowledge Integration: A Case Study in Autonomous Disassembly

The accurate estimation of 6D pose remains a challenging task within the computer vision domain, even when utilizing 3D point cloud data. Conversely, in the manufacturing domain, instances arise where leveraging prior knowledge can yield advancements in this endeavor. This study focuses on the disassembly of starter motors to augment the engineering of product life cycles. A pivotal objective in this context involves the identification and 6D pose estimation of bolts affixed to the motors, facilitating automated disassembly within the manufacturing workflow. Complicating matters, the presence of occlusions and the limitations of single-view data acquisition, notably when motors are placed in a clamping system, obscure certain portions and render some bolts imperceptible. Consequently, the development of a comprehensive pipeline capable of acquiring complete bolt information is imperative to avoid oversight in bolt detection. In this paper, employing the task of bolt detection within the scope of our project as a pertinent use case, we introduce a meticulously devised pipeline. This multi-stage pipeline effectively captures the 6D information with regard to all bolts on the motor, thereby showcasing the effective utilization of prior knowledge in handling this challenging task. The proposed methodology not only contributes to the field of 6D pose estimation but also underscores the viability of integrating domain-specific insights to tackle complex problems in manufacturing and automation.

cs.CV↗

arXiv2025

Summation of Divergent Series and Quantum Phase Transitions in Kitaev Chains with Long-Range Hopping

We study the quantum phase transitions (QPTs) in extended Kitaev chains with long-range ($1/r^α$) hopping. Formally, there are two QPT points at $μ=μ_0(α)$ and $μ_π(α)$ ($μ$ is the chemical potential) which correspond to the summations of $\sum_{m=1}^{\infty}m^{-α}$ and $\sum_{m=1}^{\infty}(-1)^{m-1}m^{-α}$, respectively. When $α\leq0$, both the series are divergent and it is usually believed that no QPTs exist. However, we find that there are two QPTs at $μ=μ_0(0)$ and $μ_π(0)$ for $α=0$ and one QPT at $μ=μ_π(α)$ for $α<0$. These QPTs are second order. The $μ_0(0)$ and $μ_π(α\leq0)$ correspond to the summations of the divergent series obtained by the analytic continuation of the Riemann $ζ$ function and Dirichlet $η$ function. Moreover, it is found that the quasiparticle energy spectra are discontinue functions of the wave vector $k$ and divide into two branches. This is quite different from that in the case of $α>0$ and induces topological phases with the winding number $ω:=\pm1/2$. At the same time, the von Neumann entropy are power law of the subchain length $L$ no matter in the gapped region or not. In addition, we also study the QPTs, topological properties, and von Neumann entropy of the systems with $α>0$.

cond-mat.stat-mech↗

arXiv2025

Probing the co-evolution of SMBHs and their hosts from scaling relations pairwise residuals: dominance of stellar velocity dispersion and host halo mass

The correlations between Supermassive Black Holes (SMBHs) and their host galaxies still defy our understanding from both the observational and theoretical perspectives. Here we perform pairwise residual analysis on the latest sample of local inactive galaxies with a uniform calibration of their photometric properties and with dynamically measured masses of their central SMBHs. The residuals reveal that stellar velocity dispersion $σ$ and, possibly host dark matter halo mass $M_{\rm halo}$, appear as the galactic properties most correlated with SMBH mass, with a secondary (weaker) correlation with spheroidal (bulge) mass $M_{\rm sph}$, as also corroborated by additional Machine Learning tests. These findings may favour energetic/kinetic feedback from Active Galactic Nuclei (AGN) as the main driver in shaping SMBH scaling relations. Two state-of-the-art hydrodynamic simulations, inclusive of kinetic AGN feedback, are able to broadly capture the mean trends observed in the residuals, although they tend to either favour $M_{\rm sph}$ as the most fundamental property, or generate too flat residuals. Increasing AGN feedback kinetic output does not improve the comparison with the data. In the Appendix we also show that the galaxies with dynamically measured SMBHs are biased high in $σ$ at fixed luminosity with respect to the full sample of local galaxies, proving that this bias is not a byproduct of stellar mass discrepancies. Overall, our results suggest that probing the SMBH-galaxy scaling relations in terms of total stellar mass alone may induce biases, and that either current data sets are incomplete, and/or that more insightful modelling is required to fully reproduce observations.

astro-ph.GA↗

arXiv2025

Shedding light on the star formation rate-halo accretion rate connection and halo quenching mechanism via DECODE, the Discrete statistical sEmi-empiriCal mODEl

Aims: The relative roles of the physical mechanisms involved in quenching galaxy star formation are still unclear. We tackle this fundamental problem with our cosmological semi-empirical model DECODE (Discrete statistical sEmi-empiriCal mODEl), designed to predict galaxy stellar mass assembly histories, from minimal input assumptions. Methods: Specifically, in this work the star formation history of each galaxy is calculated along its progenitor dark matter halo by assigning at each redshift a star formation rate extracted from a monotonic star formation rate-halo accretion rate (SFR-HAR) relation derived from abundance matching between the (observed) SFR function and the (numerically predicted) HAR function, a relation that is also predicted by the TNG100 simulation. SFRs are integrated across cosmic time to build up the mass of galaxies, which may halt their star formation following input physical quenching recipes. Results: In this work we test the popular halo quenching scenario and we find that: 1) the assumption of a monotonic relation between SFR and HAR allows to reproduce the number densities of the bulk of star-forming galaxies in the local Universe; 2) the halo quenching is sufficient to reproduce the statistics of the quenched galaxies and flat (steep) high-mass end of the SMHM relation (SMF); and 3) to align with the observed steep (flat) low-mass end of the SMHM (SMF) additional quenching processes in the least massive haloes are needed. Conclusions: DECODE is an invaluable tool and will pave the way to investigate the origin of newly observed high-redshift objects from the latest ongoing facilities such as JWST and Euclid.

astro-ph.GA↗

arXiv2025

Towards Automatic Continual Learning: A Self-Adaptive Framework for Continual Instruction Tuning

Continual instruction tuning enables large language models (LLMs) to learn incrementally while retaining past knowledge, whereas existing methods primarily focus on how to retain old knowledge rather than on selecting which new knowledge to learn. In domain-specific contexts, maintaining data quality and managing system constraints remain key challenges. To address these issues, we propose an automated continual instruction tuning framework that dynamically filters incoming data, which identify and reduce redundant data across successive updates. Our approach utilizes a small proxy model for efficient perplexity-based filtering, and updates the proxy to ensure that the filtering criteria remain aligned with the evolving state of the deployed model. Compared to existing static data selection methods, our framework can effectively handle incrementally acquired data and shifting distributions. Additionally, it addresses practical deployment challenges by enabling seamless model updates, supporting version rollback and incorporating automatic checkpoint evaluation. We evaluated the system in real-world medical scenarios. It reduced computational costs by 66.7% and improved model performance, and achieved autonomous updates, thus demonstrating its effectiveness for automatic continual instruction tuning.

cs.CL↗

arXiv2025

StringLLM: Understanding the String Processing Capability of Large Language Models

String processing, which mainly involves the analysis and manipulation of strings, is a fundamental component of modern computing. Despite the significant advancements of large language models (LLMs) in various natural language processing (NLP) tasks, their capability in string processing remains underexplored and underdeveloped. To bridge this gap, we present a comprehensive study of LLMs' string processing capability. In particular, we first propose StringLLM, a method to construct datasets for benchmarking string processing capability of LLMs. We use StringLLM to build a series of datasets, referred to as StringBench. It encompasses a wide range of string processing tasks, allowing us to systematically evaluate LLMs' performance in this area. Our evaluations indicate that LLMs struggle with accurately processing strings compared to humans. To uncover the underlying reasons for this limitation, we conduct an in-depth analysis and subsequently propose an effective approach that significantly enhances LLMs' string processing capability via fine-tuning. This work provides a foundation for future research to understand LLMs' string processing capability. Our code and data are available at https://github.com/wxl-lxw/StringLLM.

cs.CL↗

arXiv2025

Modeling the residual queue and queue-dependent capacity in a static traffic assignment problem

The residual queue during a given study period (e.g., peak hour) is an important feature that should be considered when solving a traffic assignment problem under equilibrium for strategic traffic planning. Although studies have focused extensively on static or quasi-dynamic traffic assignment models considering the residual queue, they have failed to capture the situation wherein the equilibrium link flow passing through the link is less than the link physical capacity under congested conditions. To address this critical issue, we introduce a novel static traffic assignment model that explicitly incorporates the residual queue and queue-dependent link capacity. The proposed model ensures that equilibrium link flows remain within the physical capacity bounds, yielding estimations more aligned with data observed by traffic detectors, especially in oversaturated scenarios. A generalized link cost function considering queue-dependent capacity, with an additional queuing delay term is proposed. The queuing delay term represents the added travel cost under congestion, offering a framework wherein conventional static models, both with and without physical capacity constraints, become special cases of our model. Our study rigorously analyzes the mathematical properties of the new model, establishing the theoretical uniqueness of solutions for link flow and residual queue under certain conditions. We also introduce a gradient projection-based alternating minimization algorithm tailored for the proposed model. Numerical examples are conducted to demonstrate the superiority and merit of the proposed model and solution algorithm.

eess.SY↗

arXiv2025

TextToucher: Fine-Grained Text-to-Touch Generation

Tactile sensation plays a crucial role in the development of multi-modal large models and embodied intelligence. To collect tactile data with minimal cost as possible, a series of studies have attempted to generate tactile images by vision-to-touch image translation. However, compared to text modality, visual modality-driven tactile generation cannot accurately depict human tactile sensation. In this work, we analyze the characteristics of tactile images in detail from two granularities: object-level (tactile texture, tactile shape), and sensor-level (gel status). We model these granularities of information through text descriptions and propose a fine-grained Text-to-Touch generation method (TextToucher) to generate high-quality tactile samples. Specifically, we introduce a multimodal large language model to build the text sentences about object-level tactile information and employ a set of learnable text prompts to represent the sensor-level tactile information. To better guide the tactile generation process with the built text information, we fuse the dual grains of text information and explore various dual-grain text conditioning methods within the diffusion transformer architecture. Furthermore, we propose a Contrastive Text-Touch Pre-training (CTTP) metric to precisely evaluate the quality of text-driven generated tactile data. Extensive experiments demonstrate the superiority of our TextToucher method. The source codes will be available at \url{https://github.com/TtuHamg/TextToucher}.

cs.CV↗

arXiv2024

FedCross: Intertemporal Federated Learning Under Evolutionary Games

Federated Learning (FL) mitigates privacy leakage in decentralized machine learning by allowing multiple clients to train collaboratively locally. However, dynamic mobile networks with high mobility, intermittent connectivity, and bandwidth limitation severely hinder model updates to the cloud server. Although previous studies have typically addressed user mobility issue through task reassignment or predictive modeling, frequent migrations may result in high communication overhead. Overcoming this obstacle involves not only dealing with resource constraints, but also finding ways to mitigate the challenges posed by user migrations. We therefore propose an intertemporal incentive framework, FedCross, which ensures the continuity of FL tasks by migrating interrupted training tasks to feasible mobile devices. Specifically, FedCross comprises two distinct stages. In Stage 1, we address the task allocation problem across regions under resource constraints by employing a multi-objective migration algorithm to quantify the optimal task receivers. Moreover, we adopt evolutionary game theory to capture the dynamic decision-making of users, forecasting the evolution of user proportions across different regions to mitigate frequent migrations. In Stage 2, we utilize a procurement auction mechanism to allocate rewards among base stations, ensuring that those providing high-quality models receive optimal compensation. This approach incentivizes sustained user participation, thereby ensuring the overall feasibility of FedCross. Finally, experimental results validate the theoretical soundness of FedCross and demonstrate its significant reduction in communication overhead.

cs.LG↗

arXiv2024

TRAIL: Trust-Aware Client Scheduling for Semi-Decentralized Federated Learning

Due to the sensitivity of data, Federated Learning (FL) is employed to enable distributed machine learning while safeguarding data privacy and accommodating the requirements of various devices. However, in the context of semi-decentralized FL, clients' communication and training states are dynamic. This variability arises from local training fluctuations, heterogeneous data distributions, and intermittent client participation. Most existing studies primarily focus on stable client states, neglecting the dynamic challenges inherent in real-world scenarios. To tackle this issue, we propose a TRust-Aware clIent scheduLing mechanism called TRAIL, which assesses client states and contributions, enhancing model training efficiency through selective client participation. We focus on a semi-decentralized FL framework where edge servers and clients train a shared global model using unreliable intra-cluster model aggregation and inter-cluster model consensus. First, we propose an adaptive hidden semi-Markov model to estimate clients' communication states and contributions. Next, we address a client-server association optimization problem to minimize global training loss. Using convergence analysis, we propose a greedy client scheduling algorithm. Finally, our experiments conducted on real-world datasets demonstrate that TRAIL outperforms state-of-the-art baselines, achieving an improvement of 8.7% in test accuracy and a reduction of 15.3% in training loss.

cs.LG↗

arXiv2024

Out-of-Distribution Detection with Overlap Index

Out-of-distribution (OOD) detection is crucial for the deployment of machine learning models in the open world. While existing OOD detectors are effective in identifying OOD samples that deviate significantly from in-distribution (ID) data, they often come with trade-offs. For instance, deep OOD detectors usually suffer from high computational costs, require tuning hyperparameters, and have limited interpretability, whereas traditional OOD detectors may have a low accuracy on large high-dimensional datasets. To address these limitations, we propose a novel effective OOD detection approach that employs an overlap index (OI)-based confidence score function to evaluate the likelihood of a given input belonging to the same distribution as the available ID samples. The proposed OI-based confidence score function is non-parametric, lightweight, and easy to interpret, hence providing strong flexibility and generality. Extensive empirical evaluations indicate that our OI-based OOD detector is competitive with state-of-the-art OOD detectors in terms of detection accuracy on a wide range of datasets while requiring less computation and memory costs. Lastly, we show that the proposed OI-based confidence score function inherits nice properties from OI (e.g., insensitivity to small distributional variations and robustness against Huber $ε$-contamination) and is a versatile tool for estimating OI and model accuracy in specific contexts.

cs.LG↗

arXiv2024

An Upper Bound for the Distribution Overlap Index and Its Applications

This paper proposes an easy-to-compute upper bound for the overlap index between two probability distributions without requiring any knowledge of the distribution models. The computation of our bound is time-efficient and memory-efficient and only requires finite samples. The proposed bound shows its value in one-class classification and domain shift analysis. Specifically, in one-class classification, we build a novel one-class classifier by converting the bound into a confidence score function. Unlike most one-class classifiers, the training process is not needed for our classifier. Additionally, the experimental results show that our classifier can be accurate with only a small number of in-class samples and outperform many state-of-the-art methods on various datasets in different one-class classification scenarios. In domain shift analysis, we propose a theorem based on our bound. The theorem is useful in detecting the existence of domain shift and inferring data information. The detection and inference processes are both computation-efficient and memory-efficient. Our work shows significant promise toward broadening the applications of overlap-based metrics.

cs.LG↗

arXiv2024

The synchronization of convective lifecycles in an idealized microscopic model

How a cloud ensemble responds to external forcing is a puzzle in tropical convection research. Convectively coupled gravity waves (CCGWs) in a finite domain have controllable wavelengths, providing a convenient simulation setup for studying the cloud ensemble. A multiscale analysis shows that the growth of CCGWs in a finite-domain involves not only the amplitude growth of individual clouds but also the synchronization of convective lifecycles. To understand the synchronization mechanism, we build a microscopic model with many clouds. For each cloud, the microscopic model simulates the evolution of equivalent potential temperature $θ_e$ in the boundary layer, which is reduced by convective transport and radiative cooling and increased by surface heating. At the shallow convection stage, the $θ_e$ grows until reaching an upper threshold where the convective inhibition energy is eliminated, and the system transitions to the deep convection stage. At the deep convection stage, the $θ_e$ drops until reaching a lower threshold where the convective available potential energy is exhausted, and the system transitions to the shallow convection stage. The wave influences $θ_e$ with the boundary layer convergent flow and adjusts the phase of the convective lifecycle. Numerical simulations of the microscopic model show that when the period of convection and wave equals, the wave gradually synchronizes convection. Theoretical analysis shows that the microscopic synchronization appears as the macroscopic resonant growth of the cloud ensemble. In the resonant state, the averaged $θ_e$ and vertical velocity in the boundary layer are in phase, agreeing with the cloud-permitting simulation.

physics.ao-ph↗

arXiv2024

CLIPScope: Enhancing Zero-Shot OOD Detection with Bayesian Scoring

Detection of out-of-distribution (OOD) samples is crucial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD samples without requiring in-distribution (ID) images. However, these zero-shot methods often underperform as they do not adequately consider ID class likelihoods in their detection confidence scoring. Hence, we introduce CLIPScope, a zero-shot OOD detection approach that normalizes the confidence score of a sample by class likelihoods, akin to a Bayesian posterior update. Furthermore, CLIPScope incorporates a novel strategy to mine OOD classes from a large lexical database. It selects class labels that are farthest and nearest to ID classes in terms of CLIP embedding distance to maximize coverage of OOD samples. We conduct extensive ablation studies and empirical evaluations, demonstrating state of the art performance of CLIPScope across various OOD detection benchmarks.

cs.CV↗

arXiv2024

Adaptive Domain Learning for Cross-domain Image Denoising

Different camera sensors have different noise patterns, and thus an image denoising model trained on one sensor often does not generalize well to a different sensor. One plausible solution is to collect a large dataset for each sensor for training or fine-tuning, which is inevitably time-consuming. To address this cross-domain challenge, we present a novel adaptive domain learning (ADL) scheme for cross-domain RAW image denoising by utilizing existing data from different sensors (source domain) plus a small amount of data from the new sensor (target domain). The ADL training scheme automatically removes the data in the source domain that are harmful to fine-tuning a model for the target domain (some data are harmful as adding them during training lowers the performance due to domain gaps). Also, we introduce a modulation module to adopt sensor-specific information (sensor type and ISO) to understand input data for image denoising. We conduct extensive experiments on public datasets with various smartphone and DSLR cameras, which show our proposed model outperforms prior work on cross-domain image denoising, given a small amount of image data from the target domain sensor.

cs.CV↗

arXiv2024

Combining Switching Mechanism with Re-Initialization and Anomaly Detection for Resiliency of Cyber-Physical Systems

Cyber-physical systems (CPS) play a pivotal role in numerous critical real-world applications that have stringent requirements for safety. To enhance the CPS resiliency against attacks, redundancy can be integrated in real-time controller implementations by designing strategies that switch among multiple controllers. However, existing switching strategies typically overlook remediation measures for compromised controllers, opting instead to simply exclude them. Such a solution reduces the CPS redundancy since only a subset of controllers are used. To address this gap, this work proposes a multi-controller switching strategy with periodic re-initialization to remove attacks. Controllers that finish re-initialization can be reused by the switching strategy, preserving the CPS redundancy and resiliency. The proposed switching strategy is designed to ensure that at each switching moment, a controller that has just completed re-initialization is available, minimizing the likelihood of compromise. Additionally, the controller's working period decreases with the number of involved controllers, reducing the controller's exposure time to attacks. An anomaly detector is used to detect CPS attacks during the controller's working period. Upon alarm activation, the current control signal is set to a predefined value, and a switch to an alternative controller occurs at the earliest switching moment. Our switching strategy is shown to be still effective even if the anomaly detector fails to detect (stealthy) attacks.

eess.SY↗

arXiv2024

Contrast, Imitate, Adapt: Learning Robotic Skills From Raw Human Videos

Learning robotic skills from raw human videos remains a non-trivial challenge. Previous works tackled this problem by leveraging behavior cloning or learning reward functions from videos. Despite their remarkable performances, they may introduce several issues, such as the necessity for robot actions, requirements for consistent viewpoints and similar layouts between human and robot videos, as well as low sample efficiency. To this end, our key insight is to learn task priors by contrasting videos and to learn action priors through imitating trajectories from videos, and to utilize the task priors to guide trajectories to adapt to novel scenarios. We propose a three-stage skill learning framework denoted as Contrast-Imitate-Adapt (CIA). An interaction-aware alignment transformer is proposed to learn task priors by temporally aligning video pairs. Then a trajectory generation model is used to learn action priors. To adapt to novel scenarios different from human videos, the Inversion-Interaction method is designed to initialize coarse trajectories and refine them by limited interaction. In addition, CIA introduces an optimization method based on semantic directions of trajectories for interaction security and sample efficiency. The alignment distances computed by IAAformer are used as the rewards. We evaluate CIA in six real-world everyday tasks, and empirically demonstrate that CIA significantly outperforms previous state-of-the-art works in terms of task success rate and generalization to diverse novel scenarios layouts and object instances.

cs.RO↗

arXiv2024

Rethinking Attention Module Design for Point Cloud Analysis

In recent years, there have been significant advancements in applying attention mechanisms to point cloud analysis. However, attention module variants featured in various research papers often operate under diverse settings and tasks, incorporating potential training strategies. This heterogeneity poses challenges in establishing a fair comparison among these attention module variants. In this paper, we address this issue by rethinking and exploring attention module design within a consistent base framework and settings. Both global-based and local-based attention methods are studied, with a focus on the selection basis and scales of neighbors for local-based attention. Different combinations of aggregated local features and computation methods for attention scores are evaluated, ranging from the initial addition/concatenation-based approach to the widely adopted dot product-based method and the recently proposed vector attention technique. Various position encoding methods are also investigated. Our extensive experimental analysis reveals that there is no universally optimal design across diverse point cloud tasks. Instead, drawing from best practices, we propose tailored attention modules for specific tasks, leading to superior performance on point cloud classification and segmentation benchmarks.

cs.CV↗