id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2501.02721
Learning Stochastic Nonlinear Dynamics with Embedded Latent Transfer Operators
cs.LG
We consider an operator-based latent Markov representation of a stochastic nonlinear dynamical system, where the stochastic evolution of the latent state embedded in a reproducing kernel Hilbert space is described with the corresponding transfer operator, and develop a spectral method to learn this representation based on the theory of stochastic realization. The embedding may be learned simultaneously using reproducing kernels, for example, constructed with feed-forward neural networks. We also address the generalization of sequential state-estimation (Kalman filtering) in stochastic nonlinear systems, and of operator-based eigen-mode decomposition of dynamics, for the representation. Several examples with synthetic and real-world data are shown to illustrate the empirical characteristics of our methods, and to investigate the performance of our model in sequential state-estimation and mode decomposition.
2501.02725
Artificial Intelligence in Creative Industries: Advances Prior to 2025
cs.AI
The rapid advancements in artificial intelligence (AI), particularly in generative AI and large language models (LLMs), have profoundly impacted the creative industries by enabling innovative content creation, enhancing workflows, and democratizing access to creative tools. This paper explores the significant technological shifts since our previous review in 2022, highlighting how these developments have expanded creative opportunities and efficiency. These technological advancements have enhanced the capabilities of text-to-image, text-to-video, and multimodal generation technologies. In particular, key breakthroughs in LLMs have established new benchmarks in conversational AI, while advancements in image generators have revolutionized content creation. We also discuss AI integration into post-production workflows, which has significantly accelerated and refined traditional processes. Despite these innovations, challenges remain, particularly for the media industry, due to the demands on communication traffic from creative content. We therefore include data compression and quality assessment in this paper. Furthermore, we highlight the trend toward unified AI frameworks capable of addressing multiple creative tasks and underscore the importance of human oversight to mitigate AI-generated inaccuracies. Finally, we explore AI's future potential in the creative sector, stressing the need to navigate emerging challenges to maximize its benefits while addressing associated risks.
2501.02727
Tree-based RAG-Agent Recommendation System: A Case Study in Medical Test Data
cs.IR cs.AI
We present HiRMed (Hierarchical RAG-enhanced Medical Test Recommendation), a novel tree-structured recommendation system that leverages Retrieval-Augmented Generation (RAG) for intelligent medical test recommendations. Unlike traditional vector similarity-based approaches, our system performs medical reasoning at each tree node through a specialized RAG process. Starting from the root node with initial symptoms, the system conducts step-wise medical analysis to identify potential underlying conditions and their corresponding diagnostic requirements. At each level, instead of simple matching, our RAG-enhanced nodes analyze retrieved medical knowledge to understand symptom-disease relationships and determine the most appropriate diagnostic path. The system dynamically adjusts its recommendation strategy based on medical reasoning results, considering factors such as urgency levels and diagnostic uncertainty. Experimental results demonstrate that our approach achieves superior performance in terms of coverage rate, accuracy, and miss rate compared to conventional retrieval-based methods. This work represents a significant advance in medical test recommendation by introducing medical reasoning capabilities into the traditional tree-based retrieval structure.
2501.02728
OpenGU: A Comprehensive Benchmark for Graph Unlearning
cs.LG cs.AI
Graph Machine Learning is essential for understanding and analyzing relational data. However, privacy-sensitive applications demand the ability to efficiently remove sensitive information from trained graph neural networks (GNNs), avoiding the unnecessary time and space overhead caused by retraining models from scratch. To address this issue, Graph Unlearning (GU) has emerged as a critical solution, with the potential to support dynamic graph updates in data management systems and enable scalable unlearning in distributed data systems while ensuring privacy compliance. Unlike machine unlearning in computer vision or other fields, GU faces unique difficulties due to the non-Euclidean nature of graph data and the recursive message-passing mechanism of GNNs. Additionally, the diversity of downstream tasks and the complexity of unlearning requests further amplify these challenges. Despite the proliferation of diverse GU strategies, the absence of a benchmark providing fair comparisons for GU, and the limited flexibility in combining downstream tasks and unlearning requests, have yielded inconsistencies in evaluations, hindering the development of this domain. To fill this gap, we present OpenGU, the first GU benchmark, where 16 SOTA GU algorithms and 37 multi-domain datasets are integrated, enabling various downstream tasks with 13 GNN backbones when responding to flexible unlearning requests. Based on this unified benchmark framework, we are able to provide a comprehensive and fair evaluation for GU. Through extensive experimentation, we have drawn $8$ crucial conclusions about existing GU methods, while also gaining valuable insights into their limitations, shedding light on potential avenues for future research.
2501.02732
AFed: Algorithmic Fair Federated Learning
cs.LG cs.AI
Federated Learning (FL) has gained significant attention as it facilitates collaborative machine learning among multiple clients without centralizing their data on a server. FL ensures the privacy of participating clients by locally storing their data, which creates new challenges in fairness. Traditional debiasing methods assume centralized access to sensitive information, rendering them impractical for the FL setting. Additionally, FL is more susceptible to fairness issues than centralized machine learning due to the diverse client data sources that may be associated with group information. Therefore, training a fair model in FL without access to client local data is important and challenging. This paper presents AFed, a straightforward yet effective framework for promoting group fairness in FL. The core idea is to circumvent restricted data access by learning the global data distribution. This paper proposes two approaches: AFed-G, which uses a conditional generator trained on the server side, and AFed-GAN, which improves upon AFed-G by training a conditional GAN on the client side. We augment the client data with the generated samples to help remove bias. Our theoretical analysis justifies the proposed methods, and empirical results on multiple real-world datasets demonstrate a substantial improvement in AFed over several baselines.
2501.02735
Sequence Complementor: Complementing Transformers For Time Series Forecasting with Learnable Sequences
cs.LG
Since its introduction, the transformer has shifted the development trajectory away from traditional models (e.g., RNN, MLP) in time series forecasting, which is attributed to its ability to capture global dependencies within temporal tokens. Follow-up studies have largely involved altering the tokenization and self-attention modules to better adapt Transformers for addressing special challenges like non-stationarity, channel-wise dependency, and variable correlation in time series. However, we found that the expressive capability of sequence representation is a key factor influencing Transformer performance in time forecasting after investigating several representative methods, where there is an almost linear relationship between sequence representation entropy and mean square error, with more diverse representations performing better. In this paper, we propose a novel attention mechanism with Sequence Complementors and prove feasible from an information theory perspective, where these learnable sequences are able to provide complementary information beyond current input to feed attention. We further enhance the Sequence Complementors via a diversification loss that is theoretically covered. The empirical evaluation of both long-term and short-term forecasting has confirmed its superiority over the recent state-of-the-art methods.
2501.02737
Holistic Semantic Representation for Navigational Trajectory Generation
cs.CV cs.LG
Trajectory generation has garnered significant attention from researchers in the field of spatio-temporal analysis, as it can generate substantial synthesized human mobility trajectories that enhance user privacy and alleviate data scarcity. However, existing trajectory generation methods often focus on improving trajectory generation quality from a singular perspective, lacking a comprehensive semantic understanding across various scales. Consequently, we are inspired to develop a HOlistic SEmantic Representation (HOSER) framework for navigational trajectory generation. Given an origin-and-destination (OD) pair and the starting time point of a latent trajectory, we first propose a Road Network Encoder to expand the receptive field of road- and zone-level semantics. Second, we design a Multi-Granularity Trajectory Encoder to integrate the spatio-temporal semantics of the generated trajectory at both the point and trajectory levels. Finally, we employ a Destination-Oriented Navigator to seamlessly integrate destination-oriented guidance. Extensive experiments on three real-world datasets demonstrate that HOSER outperforms state-of-the-art baselines by a significant margin. Moreover, the model's performance in few-shot learning and zero-shot learning scenarios further verifies the effectiveness of our holistic semantic representation.
2501.02738
SCSC: A Novel Standards-Compatible Semantic Communication Framework for Image Transmission
cs.IT math.IT
Joint source-channel coding (JSCC) is a promising paradigm for next-generation communication systems, particularly in challenging transmission environments. In this paper, we propose a novel standard-compatible JSCC framework for the transmission of images over multiple-input multiple-output (MIMO) channels. Different from the existing end-to-end AI-based DeepJSCC schemes, our framework consists of learnable modules that enable communication using conventional separate source and channel codes (SSCC), which makes it amenable for easy deployment on legacy systems. Specifically, the learnable modules involve a preprocessing-empowered network (PPEN) for preserving essential semantic information, and a precoder \& combiner-enhanced network (PCEN) for efficient transmission over a resource-constrained MIMO channel. We treat existing compression and channel coding modules as non-trainable blocks. Since the parameters of these modules are non-differentiable, we employ a proxy network that mimics their operations when training the learnable modules. Numerical results demonstrate that our scheme can save more than 29\% of the channel bandwidth, and requires lower complexity compared to the constrained baselines. We also show its generalization capability to unseen datasets and tasks through extensive experiments.
2501.02739
TARDiS : Text Augmentation for Refining Diversity and Separability
cs.CL cs.AI cs.LG
Text augmentation (TA) is a critical technique for text classification, especially in few-shot settings. This paper introduces a novel LLM-based TA method, TARDiS, to address challenges inherent in the generation and alignment stages of two-stage TA methods. For the generation stage, we propose two generation processes, SEG and CEG, incorporating multiple class-specific prompts to enhance diversity and separability. For the alignment stage, we introduce a class adaptation (CA) method to ensure that generated examples align with their target classes through verification and modification. Experimental results demonstrate TARDiS's effectiveness, outperforming state-of-the-art LLM-based TA methods in various few-shot text classification tasks. An in-depth analysis confirms the detailed behaviors at each stage.
2501.02740
Interpretable Recognition of Fused Magnesium Furnace Working Conditions with Deep Convolutional Stochastic Configuration Networks
cs.CV cs.AI
To address the issues of a weak generalization capability and interpretability in working condition recognition model of a fused magnesium furnace, this paper proposes an interpretable working condition recognition method based on deep convolutional stochastic configuration networks (DCSCNs). Firstly, a supervised learning mechanism is employed to generate physically meaningful Gaussian differential convolution kernels. An incremental method is utilized to construct a DCSCNs model, ensuring the convergence of recognition errors in a hierarchical manner and avoiding the iterative optimization process of convolutional kernel parameters using the widely used backpropagation algorithm. The independent coefficient of channel feature maps is defined to obtain the visualization results of feature class activation maps for the fused magnesium furnace. A joint reward function is constructed based on the recognition accuracy, the interpretable trustworthiness evaluation metrics, and the model parameter quantity. Reinforcement learning (RL) is applied to adaptively prune the convolutional kernels of the DCSCNs model, aiming to build a compact, highly performed and interpretable network. The experimental results demonstrate that the proposed method outperforms the other deep learning approaches in terms of recognition accuracy and interpretability.
2501.02741
Brick-Diffusion: Generating Long Videos with Brick-to-Wall Denoising
cs.CV
Recent advances in diffusion models have greatly improved text-driven video generation. However, training models for long video generation demands significant computational power and extensive data, leading most video diffusion models to be limited to a small number of frames. Existing training-free methods that attempt to generate long videos using pre-trained short video diffusion models often struggle with issues such as insufficient motion dynamics and degraded video fidelity. In this paper, we present Brick-Diffusion, a novel, training-free approach capable of generating long videos of arbitrary length. Our method introduces a brick-to-wall denoising strategy, where the latent is denoised in segments, with a stride applied in subsequent iterations. This process mimics the construction of a staggered brick wall, where each brick represents a denoised segment, enabling communication between frames and improving overall video quality. Through quantitative and qualitative evaluations, we demonstrate that Brick-Diffusion outperforms existing baseline methods in generating high-fidelity videos.
2501.02749
Enhancing Robot Route Optimization in Smart Logistics with Transformer and GNN Integration
cs.RO cs.AI
This research delves into advanced route optimization for robots in smart logistics, leveraging a fusion of Transformer architectures, Graph Neural Networks (GNNs), and Generative Adversarial Networks (GANs). The approach utilizes a graph-based representation encompassing geographical data, cargo allocation, and robot dynamics, addressing both spatial and resource limitations to refine route efficiency. Through extensive testing with authentic logistics datasets, the proposed method achieves notable improvements, including a 15% reduction in travel distance, a 20% boost in time efficiency, and a 10% decrease in energy consumption. These findings highlight the algorithm's effectiveness, promoting enhanced performance in intelligent logistics operations.
2501.02750
Spectrum Sharing in Satellite-Terrestrial Integrated Networks: Frameworks, Approaches, and Opportunities
eess.SP cs.IT cs.SY eess.SY math.IT
To accommodate the increasing communication needs in non-terrestrial networks (NTNs), wireless users in remote areas may require access to more spectrum than is currently allocated. Terrestrial networks (TNs), such as cellular networks, are deployed in specific areas, but many underused licensed spectrum bands remain in remote areas. Therefore, bringing NTNs to a shared spectrum with TNs can improve network capacity under reasonable interference management. However, in satellite-terrestrial integrated networks (STINs), the comprehensive coverage of a satellite and the unbalanced communication resources of STINs make it challenging to effectively manage mutual interference between NTN and TN. This article presents the fundamentals and prospects of spectrum sharing (SS) in STINs by introducing four SS frameworks, their potential application scenarios, and technical challenges. Furthermore, advanced SS approaches related to interference management in STINs and performance metrics of SS in STINs are introduced. Moreover, a preliminary performance evaluation showcases the potential for sharing the spectrum between NTN and TN. Finally, future research opportunities for SS in STINs are discussed.
2501.02751
Ultrasound-QBench: Can LLMs Aid in Quality Assessment of Ultrasound Imaging?
eess.IV cs.CV cs.MM
With the dramatic upsurge in the volume of ultrasound examinations, low-quality ultrasound imaging has gradually increased due to variations in operator proficiency and imaging circumstances, imposing a severe burden on diagnosis accuracy and even entailing the risk of restarting the diagnosis in critical cases. To assist clinicians in selecting high-quality ultrasound images and ensuring accurate diagnoses, we introduce Ultrasound-QBench, a comprehensive benchmark that systematically evaluates multimodal large language models (MLLMs) on quality assessment tasks of ultrasound images. Ultrasound-QBench establishes two datasets collected from diverse sources: IVUSQA, consisting of 7,709 images, and CardiacUltraQA, containing 3,863 images. These images encompassing common ultrasound imaging artifacts are annotated by professional ultrasound experts and classified into three quality levels: high, medium, and low. To better evaluate MLLMs, we decompose the quality assessment task into three dimensionalities: qualitative classification, quantitative scoring, and comparative assessment. The evaluation of 7 open-source MLLMs as well as 1 proprietary MLLMs demonstrates that MLLMs possess preliminary capabilities for low-level visual tasks in ultrasound image quality classification. We hope this benchmark will inspire the research community to delve deeper into uncovering and enhancing the untapped potential of MLLMs for medical imaging tasks.
2501.02754
MBTSAD: Mitigating Backdoors in Language Models Based on Token Splitting and Attention Distillation
cs.CR cs.CL
In recent years, attention-based models have excelled across various domains but remain vulnerable to backdoor attacks, often from downloading or fine-tuning on poisoned datasets. Many current methods to mitigate backdoors in NLP models rely on the pre-trained (unfine-tuned) weights, but these methods fail in scenarios where the pre-trained weights are not available. In this work, we propose MBTSAD, which can mitigate backdoors in the language model by utilizing only a small subset of clean data and does not require pre-trained weights. Specifically, MBTSAD retrains the backdoored model on a dataset generated by token splitting. Then MBTSAD leverages attention distillation, the retrained model is the teacher model, and the original backdoored model is the student model. Experimental results demonstrate that MBTSAD achieves comparable backdoor mitigation performance as the methods based on pre-trained weights while maintaining the performance on clean data. MBTSAD does not rely on pre-trained weights, enhancing its utility in scenarios where pre-trained weights are inaccessible. In addition, we simplify the min-max problem of adversarial training and visualize text representations to discover that the token splitting method in MBTSAD's first step generates Out-of-Distribution (OOD) data, leading the model to learn more generalized features and eliminate backdoor patterns.
2501.02758
Digital Twin Aided Channel Estimation: Zone-Specific Subspace Prediction and Calibration
eess.SP cs.IT math.IT
Effective channel estimation in sparse and high-dimensional environments is essential for next-generation wireless systems, particularly in large-scale MIMO deployments. This paper introduces a novel framework that leverages digital twins (DTs) as priors to enable efficient zone-specific subspace-based channel estimation (CE). Subspace-based CE significantly reduces feedback overhead by focusing on the dominant channel components, exploiting sparsity in the angular domain while preserving estimation accuracy. While DT channels may exhibit inaccuracies, their coarse-grained subspaces provide a powerful starting point, reducing the search space and accelerating convergence. The framework employs a two-step clustering process on the Grassmann manifold, combined with reinforcement learning (RL), to iteratively calibrate subspaces and align them with real-world counterparts. Simulations show that digital twins not only enable near-optimal performance but also enhance the accuracy of subspace calibration through RL, highlighting their potential as a step towards learnable digital twins.
2501.02760
CHAT: Beyond Contrastive Graph Transformer for Link Prediction in Heterogeneous Networks
cs.CE cs.LG
Link prediction in heterogeneous networks is crucial for understanding the intricacies of network structures and forecasting their future developments. Traditional methodologies often face significant obstacles, including over-smoothing-wherein the excessive aggregation of node features leads to the loss of critical structural details-and a dependency on human-defined meta-paths, which necessitate extensive domain knowledge and can be inherently restrictive. These limitations hinder the effective prediction and analysis of complex heterogeneous networks. In response to these challenges, we propose the Contrastive Heterogeneous grAph Transformer (CHAT). CHAT introduces a novel sampling-based graph transformer technique that selectively retains nodes of interest, thereby obviating the need for predefined meta-paths. The method employs an innovative connection-aware transformer to encode node sequences and their interconnections with high fidelity, guided by a dual-faceted loss function specifically designed for heterogeneous network link prediction. Additionally, CHAT incorporates an ensemble link predictor that synthesizes multiple samplings to achieve enhanced prediction accuracy. We conducted comprehensive evaluations of CHAT using three distinct drug-target interaction (DTI) datasets. The empirical results underscore CHAT's superior performance, outperforming both general-task approaches and models specialized in DTI prediction. These findings substantiate the efficacy of CHAT in addressing the complex problem of link prediction in heterogeneous networks.
2501.02761
Beyond $\mathcal{O}(\sqrt{T})$ Regret: Decoupling Learning and Decision-making in Online Linear Programming
stat.ML cs.LG math.OC
Online linear programming plays an important role in both revenue management and resource allocation, and recent research has focused on developing efficient first-order online learning algorithms. Despite the empirical success of first-order methods, they typically achieve a regret no better than $\mathcal{O} ( \sqrt{T} )$, which is suboptimal compared to the $\mathcal{O} (\log T)$ bound guaranteed by the state-of-the-art linear programming (LP)-based online algorithms. This paper establishes a general framework that improves upon the $\mathcal{O} ( \sqrt{T} )$ result when the LP dual problem exhibits certain error bound conditions. For the first time, we show that first-order learning algorithms achieve $o( \sqrt{T} )$ regret in the continuous support setting and $\mathcal{O} (\log T)$ regret in the finite support setting beyond the non-degeneracy assumption. Our results significantly improve the state-of-the-art regret results and provide new insights for sequential decision-making.
2501.02763
LDMapNet-U: An End-to-End System for City-Scale Lane-Level Map Updating
cs.CV
An up-to-date city-scale lane-level map is an indispensable infrastructure and a key enabling technology for ensuring the safety and user experience of autonomous driving systems. In industrial scenarios, reliance on manual annotation for map updates creates a critical bottleneck. Lane-level updates require precise change information and must ensure consistency with adjacent data while adhering to strict standards. Traditional methods utilize a three-stage approach-construction, change detection, and updating-which often necessitates manual verification due to accuracy limitations. This results in labor-intensive processes and hampers timely updates. To address these challenges, we propose LDMapNet-U, which implements a new end-to-end paradigm for city-scale lane-level map updating. By reconceptualizing the update task as an end-to-end map generation process grounded in historical map data, we introduce a paradigm shift in map updating that simultaneously generates vectorized maps and change information. To achieve this, a Prior-Map Encoding (PME) module is introduced to effectively encode historical maps, serving as a critical reference for detecting changes. Additionally, we incorporate a novel Instance Change Prediction (ICP) module that learns to predict associations with historical maps. Consequently, LDMapNet-U simultaneously achieves vectorized map element generation and change detection. To demonstrate the superiority and effectiveness of LDMapNet-U, extensive experiments are conducted using large-scale real-world datasets. In addition, LDMapNet-U has been successfully deployed in production at Baidu Maps since April 2024, supporting map updating for over 360 cities and significantly shortening the update cycle from quarterly to weekly. The updated maps serve hundreds of millions of users and are integrated into the autonomous driving systems of several leading vehicle companies.
2501.02765
Visual Large Language Models for Generalized and Specialized Applications
cs.CV cs.AI
Visual-language models (VLM) have emerged as a powerful tool for learning a unified embedding space for vision and language. Inspired by large language models, which have demonstrated strong reasoning and multi-task capabilities, visual large language models (VLLMs) are gaining increasing attention for building general-purpose VLMs. Despite the significant progress made in VLLMs, the related literature remains limited, particularly from a comprehensive application perspective, encompassing generalized and specialized applications across vision (image, video, depth), action, and language modalities. In this survey, we focus on the diverse applications of VLLMs, examining their using scenarios, identifying ethics consideration and challenges, and discussing future directions for their development. By synthesizing these contents, we aim to provide a comprehensive guide that will pave the way for future innovations and broader applications of VLLMs. The paper list repository is available: https://github.com/JackYFL/awesome-VLLMs.
2501.02766
Are GNNs Effective for Multimodal Fault Diagnosis in Microservice Systems?
cs.SE cs.AI
Fault diagnosis in microservice systems has increasingly embraced multimodal observation data for a holistic and multifaceted view of the system, with Graph Neural Networks (GNNs) commonly employed to model complex service dependencies. However, despite the intuitive appeal, there remains a lack of compelling justification for the adoption of GNNs, as no direct evidence supports their necessity or effectiveness. To critically evaluate the current use of GNNs, we propose DiagMLP, a simple topology-agnostic baseline as a substitute for GNNs in fault diagnosis frameworks. Through experiments on five public datasets, we surprisingly find that DiagMLP performs competitively with and even outperforms GNN-based methods in fault diagnosis tasks, indicating that the current paradigm of using GNNs to model service dependencies has not yet demonstrated a tangible contribution. We further discuss potential reasons for this observation and advocate shifting the focus from solely pursuing novel model designs to developing challenging datasets, standardizing preprocessing protocols, and critically evaluating the utility of advanced deep learning modules.
2501.02767
Enhancing Trustworthiness of Graph Neural Networks with Rank-Based Conformal Training
cs.LG cs.AI
Graph Neural Networks (GNNs) has been widely used in a variety of fields because of their great potential in representing graph-structured data. However, lacking of rigorous uncertainty estimations limits their application in high-stakes. Conformal Prediction (CP) can produce statistically guaranteed uncertainty estimates by using the classifier's probability estimates to obtain prediction sets, which contains the true class with a user-specified probability. In this paper, we propose a Rank-based CP during training framework to GNNs (RCP-GNN) for reliable uncertainty estimates to enhance the trustworthiness of GNNs in the node classification scenario. By exploiting rank information of the classifier's outcome, prediction sets with desired coverage rate can be efficiently constructed. The strategy of CP during training with differentiable rank-based conformity loss function is further explored to adapt prediction sets according to network topology information. In this way, the composition of prediction sets can be guided by the goal of jointly reducing inefficiency and probability estimation errors. Extensive experiments on several real-world datasets show that our model achieves any pre-defined target marginal coverage while significantly reducing the inefficiency compared with state-of-the-art methods.
2501.02770
Multi-Agent Path Finding under Limited Communication Range Constraint via Dynamic Leading
cs.AI cs.MA cs.RO
This paper proposes a novel framework to handle a multi-agent path finding problem under a limited communication range constraint, where all agents must have a connected communication channel to the rest of the team. Many existing approaches to multi-agent path finding (e.g., leader-follower platooning) overcome computational challenges of planning in this domain by planning one agent at a time in a fixed order. However, fixed leader-follower approaches can become stuck during planning, limiting their practical utility in dense-clutter environments. To overcome this limitation, we develop dynamic leading multi-agent path finding, which allows for dynamic reselection of the leading agent during path planning whenever progress cannot be made. The experiments show the efficiency of our framework, which can handle up to 25 agents with more than 90% success-rate across five environment types where baselines routinely fail.
2501.02771
WorldPose: A World Cup Dataset for Global 3D Human Pose Estimation
cs.CV
We present WorldPose, a novel dataset for advancing research in multi-person global pose estimation in the wild, featuring footage from the 2022 FIFA World Cup. While previous datasets have primarily focused on local poses, often limited to a single person or in constrained, indoor settings, the infrastructure deployed for this sporting event allows access to multiple fixed and moving cameras in different stadiums. We exploit the static multi-view setup of HD cameras to recover the 3D player poses and motions with unprecedented accuracy given capture areas of more than 1.75 acres. We then leverage the captured players' motions and field markings to calibrate a moving broadcasting camera. The resulting dataset comprises more than 80 sequences with approx 2.5 million 3D poses and a total traveling distance of over 120 km. Subsequently, we conduct an in-depth analysis of the SOTA methods for global pose estimation. Our experiments demonstrate that WorldPose challenges existing multi-person techniques, supporting the potential for new research in this area and others, such as sports analysis. All pose annotations (in SMPL format), broadcasting camera parameters and footage will be released for academic research purposes.
2501.02772
GeAR: Generation Augmented Retrieval
cs.IR cs.CL
Document retrieval techniques form the foundation for the development of large-scale information systems. The prevailing methodology is to construct a bi-encoder and compute the semantic similarity. However, such scalar similarity is difficult to reflect enough information and impedes our comprehension of the retrieval results. In addition, this computational process mainly emphasizes the global semantics and ignores the fine-grained semantic relationship between the query and the complex text in the document. In this paper, we propose a new method called $\textbf{Ge}$neration $\textbf{A}$ugmented $\textbf{R}$etrieval ($\textbf{GeAR}$) that incorporates well-designed fusion and decoding modules. This enables GeAR to generate the relevant text from documents based on the fused representation of the query and the document, thus learning to "focus on" the fine-grained information. Also when used as a retriever, GeAR does not add any computational burden over bi-encoders. To support the training of the new framework, we have introduced a pipeline to efficiently synthesize high-quality data by utilizing large language models. GeAR exhibits competitive retrieval and localization performance across diverse scenarios and datasets. Moreover, the qualitative analysis and the results generated by GeAR provide novel insights into the interpretation of retrieval results. The code, data, and models will be released after completing technical review to facilitate future research.
2501.02773
Unsupervised Domain Adaptation for Occlusion Resilient Human Pose Estimation
cs.CV
Occlusions are a significant challenge to human pose estimation algorithms, often resulting in inaccurate and anatomically implausible poses. Although current occlusion-robust human pose estimation algorithms exhibit impressive performance on existing datasets, their success is largely attributed to supervised training and the availability of additional information, such as multiple views or temporal continuity. Furthermore, these algorithms typically suffer from performance degradation under distribution shifts. While existing domain adaptive human pose estimation algorithms address this bottleneck, they tend to perform suboptimally when the target domain images are occluded, a common occurrence in real-life scenarios. To address these challenges, we propose OR-POSE: Unsupervised Domain Adaptation for Occlusion Resilient Human POSE Estimation. OR-POSE is an innovative unsupervised domain adaptation algorithm which effectively mitigates domain shifts and overcomes occlusion challenges by employing the mean teacher framework for iterative pseudo-label refinement. Additionally, OR-POSE reinforces realistic pose prediction by leveraging a learned human pose prior which incorporates the anatomical constraints of humans in the adaptation process. Lastly, OR-POSE avoids overfitting to inaccurate pseudo labels generated from heavily occluded images by employing a novel visibility-based curriculum learning approach. This enables the model to gradually transition from training samples with relatively less occlusion to more challenging, heavily occluded samples. Extensive experiments show that OR-POSE outperforms existing analogous state-of-the-art algorithms by $\sim$ 7% on challenging occluded human pose estimation datasets.
2501.02774
Learn A Flexible Exploration Model for Parameterized Action Markov Decision Processes
cs.LG
Hybrid action models are widely considered an effective approach to reinforcement learning (RL) modeling. The current mainstream method is to train agents under Parameterized Action Markov Decision Processes (PAMDPs), which performs well in specific environments. Unfortunately, these models either exhibit drastic low learning efficiency in complex PAMDPs or lose crucial information in the conversion between raw space and latent space. To enhance the learning efficiency and asymptotic performance of the agent, we propose a model-based RL (MBRL) algorithm, FLEXplore. FLEXplore learns a parameterized-action-conditioned dynamics model and employs a modified Model Predictive Path Integral control. Unlike conventional MBRL algorithms, we carefully design the dynamics loss function and reward smoothing process to learn a loose yet flexible model. Additionally, we use the variational lower bound to maximize the mutual information between the state and the hybrid action, enhancing the exploration effectiveness of the agent. We theoretically demonstrate that FLEXplore can reduce the regret of the rollout trajectory through the Wasserstein Metric under given Lipschitz conditions. Our empirical results on several standard benchmarks show that FLEXplore has outstanding learning efficiency and asymptotic performance compared to other baselines.
2501.02778
ICFNet: Integrated Cross-modal Fusion Network for Survival Prediction
eess.IV cs.AI cs.CV
Survival prediction is a crucial task in the medical field and is essential for optimizing treatment options and resource allocation. However, current methods often rely on limited data modalities, resulting in suboptimal performance. In this paper, we propose an Integrated Cross-modal Fusion Network (ICFNet) that integrates histopathology whole slide images, genomic expression profiles, patient demographics, and treatment protocols. Specifically, three types of encoders, a residual orthogonal decomposition module and a unification fusion module are employed to merge multi-modal features to enhance prediction accuracy. Additionally, a balanced negative log-likelihood loss function is designed to ensure fair training across different patients. Extensive experiments demonstrate that our ICFNet outperforms state-of-the-art algorithms on five public TCGA datasets, including BLCA, BRCA, GBMLGG, LUAD, and UCEC, and shows its potential to support clinical decision-making and advance precision medicine. The codes are available at: https://github.com/binging512/ICFNet.
2501.02781
From Dense to Sparse: Event Response for Enhanced Residential Load Forecasting
cs.LG
Residential load forecasting (RLF) is crucial for resource scheduling in power systems. Most existing methods utilize all given load records (dense data) to indiscriminately extract the dependencies between historical and future time series. However, there exist important regular patterns residing in the event-related associations among different appliances (sparse knowledge), which have yet been ignored. In this paper, we propose an Event-Response Knowledge Guided approach (ERKG) for RLF by incorporating the estimation of electricity usage events for different appliances, mining event-related sparse knowledge from the load series. With ERKG, the event-response estimation enables portraying the electricity consumption behaviors of residents, revealing regular variations in appliance operational states. To be specific, ERKG consists of knowledge extraction and guidance: i) a forecasting model is designed for the electricity usage events by estimating appliance operational states, aiming to extract the event-related sparse knowledge; ii) a novel knowledge-guided mechanism is established by fusing such state estimates of the appliance events into the RLF model, which can give particular focuses on the patterns of users' electricity consumption behaviors. Notably, ERKG can flexibly serve as a plug-in module to boost the capability of existing forecasting models by leveraging event response. In numerical experiments, extensive comparisons and ablation studies have verified the effectiveness of our ERKG, e.g., over 8% MAE can be reduced on the tested state-of-the-art forecasting models.
2501.02785
Hybrid deep convolution model for lung cancer detection with transfer learning
cs.CV cs.AI cs.LG
Advances in healthcare research have significantly enhanced our understanding of disease mechanisms, diagnostic precision, and therapeutic options. Yet, lung cancer remains one of the leading causes of cancer-related mortality worldwide due to challenges in early and accurate diagnosis. While current lung cancer detection models show promise, there is considerable potential for further improving the accuracy for timely intervention. To address this challenge, we introduce a hybrid deep convolution model leveraging transfer learning, named the Maximum Sensitivity Neural Network (MSNN). MSNN is designed to improve the precision of lung cancer detection by refining sensitivity and specificity. This model has surpassed existing deep learning approaches through experimental validation, achieving an accuracy of 98% and a sensitivity of 97%. By overlaying sensitivity maps onto lung Computed Tomography (CT) scans, it enables the visualization of regions most indicative of malignant or benign classifications. This innovative method demonstrates exceptional performance in distinguishing lung cancer with minimal false positives, thereby enhancing the accuracy of medical diagnoses.
2501.02786
CCStereo: Audio-Visual Contextual and Contrastive Learning for Binaural Audio Generation
cs.SD cs.CV eess.AS
Binaural audio generation (BAG) aims to convert monaural audio to stereo audio using visual prompts, requiring a deep understanding of spatial and semantic information. However, current models risk overfitting to room environments and lose fine-grained spatial details. In this paper, we propose a new audio-visual binaural generation model incorporating an audio-visual conditional normalisation layer that dynamically aligns the mean and variance of the target difference audio features using visual context, along with a new contrastive learning method to enhance spatial sensitivity by mining negative samples from shuffled visual features. We also introduce a cost-efficient way to utilise test-time augmentation in video data to enhance performance. Our approach achieves state-of-the-art generation accuracy on the FAIR-Play and MUSIC-Stereo benchmarks.
2501.02788
GLoG-CSUnet: Enhancing Vision Transformers with Adaptable Radiomic Features for Medical Image Segmentation
cs.CV cs.AI cs.LG
Vision Transformers (ViTs) have shown promise in medical image semantic segmentation (MISS) by capturing long-range correlations. However, ViTs often struggle to model local spatial information effectively, which is essential for accurately segmenting fine anatomical details, particularly when applied to small datasets without extensive pre-training. We introduce Gabor and Laplacian of Gaussian Convolutional Swin Network (GLoG-CSUnet), a novel architecture enhancing Transformer-based models by incorporating learnable radiomic features. This approach integrates dynamically adaptive Gabor and Laplacian of Gaussian (LoG) filters to capture texture, edge, and boundary information, enhancing the feature representation processed by the Transformer model. Our method uniquely combines the long-range dependency modeling of Transformers with the texture analysis capabilities of Gabor and LoG features. Evaluated on the Synapse multi-organ and ACDC cardiac segmentation datasets, GLoG-CSUnet demonstrates significant improvements over state-of-the-art models, achieving a 1.14% increase in Dice score for Synapse and 0.99% for ACDC, with minimal computational overhead (only 15 and 30 additional parameters, respectively). GLoG-CSUnet's flexible design allows integration with various base models, offering a promising approach for incorporating radiomics-inspired feature extraction in Transformer architectures for medical image analysis. The code implementation is available on GitHub at: https://github.com/HAAIL/GLoG-CSUnet.
2501.02790
Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model
cs.CL cs.AI
Reinforcement learning from human feedback (RLHF) has been widely adopted to align language models (LMs) with human preference. Prior RLHF works typically take a bandit formulation, which, though intuitive, ignores the sequential nature of LM generation and can suffer from the sparse reward issue. While recent works propose dense token-level RLHF, treating each token as an action may be oversubtle to proper reward assignment. In this paper, we seek to get the best of both by training and utilizing a segment-level reward model, which assigns a reward to each semantically complete text segment that spans over a short sequence of tokens. For reward learning, our method allows dynamic text segmentation and compatibility with standard sequence-preference datasets. For effective RL-based LM training against segment reward, we generalize the classical scalar bandit reward normalizers into location-aware normalizer functions and interpolate the segment reward for further densification. With these designs, our method performs competitively on three popular RLHF benchmarks for LM policy: AlpacaEval 2.0, Arena-Hard, and MT-Bench. Ablation studies are conducted to further demonstrate our method.
2501.02791
Orthogonal greedy algorithm for linear operator learning with shallow neural network
math.NA cs.LG cs.NA
Greedy algorithms, particularly the orthogonal greedy algorithm (OGA), have proven effective in training shallow neural networks for fitting functions and solving partial differential equations (PDEs). In this paper, we extend the application of OGA to the tasks of linear operator learning, which is equivalent to learning the kernel function through integral transforms. Firstly, a novel greedy algorithm is developed for kernel estimation rate in a new semi-inner product, which can be utilized to approximate the Green's function of linear PDEs from data. Secondly, we introduce the OGA for point-wise kernel estimation to further improve the approximation rate, achieving orders of accuracy improvement across various tasks and baseline models. In addition, we provide a theoretical analysis on the kernel estimation problem and the optimal approximation rates for both algorithms, establishing their efficacy and potential for future applications in PDEs and operator learning tasks.
2501.02792
Gaming on Coincident Peak Shaving: Equilibrium and Strategic Behavior
eess.SY cs.SY
Coincident peak demand charges are imposed by power system operators or electric utilities when the overall system demand, aggregated across multiple consumers, reaches its peak. These charges incentivize consumers to reduce their demand during peak periods, a practice known as coincident peak shaving. In this paper, we analyze the coincident peak shaving problem through the lens of game theory, developing a theoretical model to examine the impact of strategic consumer behavior on system efficiency. We demonstrate that the game structure exhibits varying characteristics - concave, quasiconcave/discontinuous, or non-concave/discontinuous - depending on the extent of consumers demand-shifting capabilities. For a two-agent, two-period setting, we derive closed-form Nash equilibrium solutions under each condition and generalize our findings to cases with multiple agents. We prove the stability of the equilibrium points and present an algorithm for computing equilibrium outcomes across all game scenarios. We also show that the peak-shaving effectiveness of the game model matches that of the centralized peak-shaving model but with increased levels of anarchy. In the cases of quasiconcave and non-concave game conditions, we analytically demonstrate in the two-agent setting that anarchy increases with consumers' flexibility and inequity, as measured by their marginal shifting costs, and we also analyze the influence of the number of agents on anarchy. Finally, we provide numerical simulations to validate our theoretical results.
2501.02793
Fairness Through Matching
cs.AI cs.LG stat.ML
Group fairness requires that different protected groups, characterized by a given sensitive attribute, receive equal outcomes overall. Typically, the level of group fairness is measured by the statistical gap between predictions from different protected groups. In this study, we reveal an implicit property of existing group fairness measures, which provides an insight into how the group-fair models behave. Then, we develop a new group-fair constraint based on this implicit property to learn group-fair models. To do so, we first introduce a notable theoretical observation: every group-fair model has an implicitly corresponding transport map between the input spaces of each protected group. Based on this observation, we introduce a new group fairness measure termed Matched Demographic Parity (MDP), which quantifies the averaged gap between predictions of two individuals (from different protected groups) matched by a given transport map. Then, we prove that any transport map can be used in MDP to learn group-fair models, and develop a novel algorithm called Fairness Through Matching (FTM), which learns a group-fair model using MDP constraint with an user-specified transport map. We specifically propose two favorable types of transport maps for MDP, based on the optimal transport theory, and discuss their advantages. Experiments reveal that FTM successfully trains group-fair models with certain desirable properties by choosing the transport map accordingly.
2501.02795
InfiFusion: A Unified Framework for Enhanced Cross-Model Reasoning via LLM Fusion
cs.CL cs.CV
We introduce InfiFusion, an efficient training pipeline designed to integrate multiple domain-specialized Large Language Models (LLMs) into a single pivot model, effectively harnessing the strengths of each source model. Traditional fusion methods either merge model parameters directly or rely on knowledge distillation with rigid assumptions, limiting their flexibility and efficiency. InfiFusion overcomes these limitations by enhancing Universal Logit Distillation (ULD) with Top-K selection and Logits Standardization. We propose two fusion strategies: Pairwise Fusion (InfiFusion$_p$), where each source model knowledge is distilled individually into the pivot model followed by merging and Unified Fusion (InfiFusion$_u$), where knowledge from all source models is distilled simultaneously into the pivot model. InfiFusion outperforms the state-of-the-art models, such as Qwen-2.5-14B-Instruct and Phi-4, across 11 widely applied benchmarks covering reasoning, coding, mathematics, and instruction-following tasks. Notably, InfiFusion achieves this superior performance while significantly reduces computational costs, completing full training with only 160 H800 GPU hours compared to the millions typically required for traditional LLM training.
2501.02796
GraphDART: Graph Distillation for Efficient Advanced Persistent Threat Detection
cs.CR cs.LG
Cyber-physical-social systems (CPSSs) have emerged in many applications over recent decades, requiring increased attention to security concerns. The rise of sophisticated threats like Advanced Persistent Threats (APTs) makes ensuring security in CPSSs particularly challenging. Provenance graph analysis has proven effective for tracing and detecting anomalies within systems, but the sheer size and complexity of these graphs hinder the efficiency of existing methods, especially those relying on graph neural networks (GNNs). To address these challenges, we present GraphDART, a modular framework designed to distill provenance graphs into compact yet informative representations, enabling scalable and effective anomaly detection. GraphDART can take advantage of diverse graph distillation techniques, including classic and modern graph distillation methods, to condense large provenance graphs while preserving essential structural and contextual information. This approach significantly reduces computational overhead, allowing GNNs to learn from distilled graphs efficiently and enhance detection performance. Extensive evaluations on benchmark datasets demonstrate the robustness of GraphDART in detecting malicious activities across cyber-physical-social systems. By optimizing computational efficiency, GraphDART provides a scalable and practical solution to safeguard interconnected environments against APTs.
2501.02798
Ray-Tracing Channel Modeling for LEO Satellite-to-Ground Communication Systems
eess.SY cs.SY
Based on the vision of global coverage for sixth-generation (6G) wireless communication systems, the low earth orbit (LEO) satellite-to-ground channel model for urban scenarios has emerged as highly important for the system design. In this paper, we propose an LEO satellite-to-ground channel model through shooting and bouncing rays (SBR) algorithm to analyze the channel characteristics. The orbit of LEO is modeled by the simplified general perturbations 4 (SGP4), and an accurate celestial model is applied to calculate the Doppler shift of multipath in a transmission time window of LEO satellite-to-ground communications. Channel characteristics of LEO satellite-to-ground communications such as the root-mean-square (RMS) delay spread, the Doppler shift, and the received power at different times are obtained. The simulation results show that the received power is only significantly noticeable in the transmission time window when the satellite is close to the receiver. Proposed model validates the effectiveness of ray-tracing in actual LEO satellite-to-ground communication scenarios and extends the calculation of the Doppler shift.
2501.02800
COph100: A comprehensive fundus image registration dataset from infants constituting the "RIDIRP" database
cs.CV cs.CE
Retinal image registration is vital for diagnostic therapeutic applications within the field of ophthalmology. Existing public datasets, focusing on adult retinal pathologies with high-quality images, have limited number of image pairs and neglect clinical challenges. To address this gap, we introduce COph100, a novel and challenging dataset known as the Comprehensive Ophthalmology Retinal Image Registration dataset for infants with a wide range of image quality issues constituting the public "RIDIRP" database. COph100 consists of 100 eyes, each with 2 to 9 examination sessions, amounting to a total of 491 image pairs carefully selected from the publicly available dataset. We manually labeled the corresponding ground truth image points and provided automatic vessel segmentation masks for each image. We have assessed COph100 in terms of image quality and registration outcomes using state-of-the-art algorithms. This resource enables a robust comparison of retinal registration methodologies and aids in the analysis of disease progression in infants, thereby deepening our understanding of pediatric ophthalmic conditions.
2501.02803
Enhancing Lifelong Multi-Agent Path Finding with Cache Mechanism
cs.RO cs.AI
Multi-Agent Path Finding (MAPF), which focuses on finding collision-free paths for multiple robots, is crucial in autonomous warehouse operations. Lifelong MAPF (L-MAPF), where agents are continuously reassigned new targets upon completing their current tasks, offers a more realistic approximation of real-world warehouse scenarios. While cache storage systems can enhance efficiency and reduce operational costs, existing approaches primarily rely on expectations and mathematical models, often without adequately addressing the challenges of multi-robot planning and execution. In this paper, we introduce a novel mechanism called Lifelong MAPF with Cache Mechanism (L-MAPF-CM), which integrates high-level cache storage with low-level path planning. We have involved a new type of map grid called cache for temporary item storage. Additionally, we involved a task assigner (TA) with a locking mechanism to bridge the gap between the new cache grid and L-MAPF algorithm. The TA dynamically allocates target locations to agents based on their status in various scenarios. We evaluated L-MAPF-CM using different cache replacement policies and task distributions. L-MAPF-CM has demonstrated performance improvements particularly with high cache hit rates and smooth traffic conditions.
2501.02807
AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scene
cs.CV
Compared to frame-based methods, computational neuromorphic imaging using event cameras offers significant advantages, such as minimal motion blur, enhanced temporal resolution, and high dynamic range. The multi-view consistency of Neural Radiance Fields combined with the unique benefits of event cameras, has spurred recent research into reconstructing NeRF from data captured by moving event cameras. While showing impressive performance, existing methods rely on ideal conditions with the availability of uniform and high-quality event sequences and accurate camera poses, and mainly focus on the object level reconstruction, thus limiting their practical applications. In this work, we propose AE-NeRF to address the challenges of learning event-based NeRF from non-ideal conditions, including non-uniform event sequences, noisy poses, and various scales of scenes. Our method exploits the density of event streams and jointly learn a pose correction module with an event-based NeRF (e-NeRF) framework for robust 3D reconstruction from inaccurate camera poses. To generalize to larger scenes, we propose hierarchical event distillation with a proposal e-NeRF network and a vanilla e-NeRF network to resample and refine the reconstruction process. We further propose an event reconstruction loss and a temporal loss to improve the view consistency of the reconstructed scene. We established a comprehensive benchmark that includes large-scale scenes to simulate practical non-ideal conditions, incorporating both synthetic and challenging real-world event datasets. The experimental results show that our method achieves a new state-of-the-art in event-based 3D reconstruction.
2501.02808
DarkFarseer: Inductive Spatio-temporal Kriging via Hidden Style Enhancement and Sparsity-Noise Mitigation
cs.LG
With the rapid growth of the Internet of Things and Cyber-Physical Systems, widespread sensor deployment has become essential. However, the high costs of building sensor networks limit their scale and coverage, making fine-grained deployment challenging. Inductive Spatio-Temporal Kriging (ISK) addresses this issue by introducing virtual sensors. Based on graph neural networks (GNNs) extracting the relationships between physical and virtual sensors, ISK can infer the measurements of virtual sensors from physical sensors. However, current ISK methods rely on conventional message-passing mechanisms and network architectures, without effectively extracting spatio-temporal features of physical sensors and focusing on representing virtual sensors. Additionally, existing graph construction methods face issues of sparse and noisy connections, destroying ISK performance. To address these issues, we propose DarkFarseer, a novel ISK framework with three key components. First, we propose the Neighbor Hidden Style Enhancement module with a style transfer strategy to enhance the representation of virtual nodes in a temporal-then-spatial manner to better extract the spatial relationships between physical and virtual nodes. Second, we propose Virtual-Component Contrastive Learning, which aims to enrich the node representation by establishing the association between the patterns of virtual nodes and the regional patterns within graph components. Lastly, we design a Similarity-Based Graph Denoising Strategy, which reduces the connectivity strength of noisy connections around virtual nodes and their neighbors based on their temporal information and regional spatial patterns. Extensive experiments demonstrate that DarkFarseer significantly outperforms existing ISK methods.
2501.02809
Theoretical Data-Driven MobilePosenet: Lightweight Neural Network for Accurate Calibration-Free 5-DOF Magnet Localization
cs.RO
Permanent magnet tracking using the external sensor array is crucial for the accurate localization of wireless capsule endoscope robots. Traditional tracking algorithms, based on the magnetic dipole model and Levenberg-Marquardt (LM) algorithm, face challenges related to computational delays and the need for initial position estimation. More recently proposed neural network-based approaches often require extensive hardware calibration and real-world data collection, which are time-consuming and labor-intensive. To address these challenges, we propose MobilePosenet, a lightweight neural network architecture that leverages depthwise separable convolutions to minimize computational cost and a channel attention mechanism to enhance localization accuracy. Besides, the inputs to the network integrate the sensors' coordinate information and random noise, compensating for the discrepancies between the theoretical model and the actual magnetic fields and thus allowing MobilePosenet to be trained entirely on theoretical data. Experimental evaluations conducted in a \(90 \times 90 \times 80\) mm workspace demonstrate that MobilePosenet exhibits excellent 5-DOF localization accuracy ($1.54 \pm 1.03$ mm and $2.24 \pm 1.84^{\circ}$) and inference speed (0.9 ms) against state-of-the-art methods trained on real-world data. Since network training relies solely on theoretical data, MobilePosenet can eliminate the hardware calibration and real-world data collection process, improving the generalizability of this permanent magnet localization method and the potential for rapid adoption in different clinical settings.
2501.02811
First-place Solution for Streetscape Shop Sign Recognition Competition
cs.CV
Text recognition technology applied to street-view storefront signs is increasingly utilized across various practical domains, including map navigation, smart city planning analysis, and business value assessments in commercial districts. This technology holds significant research and commercial potential. Nevertheless, it faces numerous challenges. Street view images often contain signboards with complex designs and diverse text styles, complicating the text recognition process. A notable advancement in this field was introduced by our team in a recent competition. We developed a novel multistage approach that integrates multimodal feature fusion, extensive self-supervised training, and a Transformer-based large model. Furthermore, innovative techniques such as BoxDQN, which relies on reinforcement learning, and text rectification methods were employed, leading to impressive outcomes. Comprehensive experiments have validated the effectiveness of these methods, showcasing our potential to enhance text recognition capabilities in complex urban environments.
2501.02814
Analogue Forecast System for Daily Precipitation Prediction Using Autoencoder Feature Extraction: Application in Hong Kong
physics.ao-ph cs.LG
In the Hong Kong Observatory, the Analogue Forecast System (AFS) for precipitation has been providing useful reference in predicting possible daily rainfall scenarios for the next 9 days, by identifying historical cases with similar weather patterns to the latest output from the deterministic model of the European Centre for Medium-Range Weather Forecasts (ECMWF). Recent advances in machine learning allow more sophisticated models to be trained using historical data and the patterns of high-impact weather events to be represented more effectively. As such, an enhanced AFS has been developed using the deep learning technique autoencoder. The datasets of the fifth generation of the ECMWF Reanalysis (ERA5) are utilised where more meteorological elements in higher horizontal, vertical and temporal resolutions are available as compared to the previous ECMWF reanalysis products used in the existing AFS. The enhanced AFS features four major steps in generating the daily rain class forecasts: (1) preprocessing of gridded ERA5 and ECMWF model forecast, (2) feature extraction by the pretrained autoencoder, (3) application of optimised feature weightings based on historical cases, and (4) calculation of the final rain class from a weighted ensemble of top analogues. The enhanced AFS demonstrates a consistent and superior performance over the existing AFS, especially in capturing heavy rain cases, during the verification period from 2019 to 2022. This paper presents the detailed formulation of the enhanced AFS and discusses its advantages and limitations in supporting precipitation forecasting in Hong Kong.
2501.02815
Local Reactive Control for Mobile Manipulators with Whole-Body Safety in Complex Environments
cs.RO cs.SY eess.SY
Mobile manipulators typically encounter significant challenges in navigating narrow, cluttered environments due to their high-dimensional state spaces and complex kinematics. While reactive methods excel in dynamic settings, they struggle to efficiently incorporate complex, coupled constraints across the entire state space. In this work, we present a novel local reactive controller that reformulates the time-domain single-step problem into a multi-step optimization problem in the spatial domain, leveraging the propagation of a serial kinematic chain. This transformation facilitates the formulation of customized, decoupled link-specific constraints, which is further solved efficiently with augmented Lagrangian differential dynamic programming (AL-DDP). Our approach naturally absorbs spatial kinematic propagation in the forward pass and processes all link-specific constraints simultaneously during the backward pass, enhancing both constraint management and computational efficiency. Notably, in this framework, we formulate collision avoidance constraints for each link using accurate geometric models with extracted free regions, and this improves the maneuverability of the mobile manipulator in narrow, cluttered spaces. Experimental results showcase significant improvements in safety, efficiency, and task completion rates. These findings underscore the robustness of the proposed method, particularly in narrow, cluttered environments where conventional approaches could falter. The open-source project can be found at https://github.com/Chunx1nZHENG/MM-with-Whole-Body-Safety-Release.git.
2501.02816
InpDiffusion: Image Inpainting Localization via Conditional Diffusion Models
cs.CV cs.AI
As artificial intelligence advances rapidly, particularly with the advent of GANs and diffusion models, the accuracy of Image Inpainting Localization (IIL) has become increasingly challenging. Current IIL methods face two main challenges: a tendency towards overconfidence, leading to incorrect predictions; and difficulty in detecting subtle tampering boundaries in inpainted images. In response, we propose a new paradigm that treats IIL as a conditional mask generation task utilizing diffusion models. Our method, InpDiffusion, utilizes the denoising process enhanced by the integration of image semantic conditions to progressively refine predictions. During denoising, we employ edge conditions and introduce a novel edge supervision strategy to enhance the model's perception of edge details in inpainted objects. Balancing the diffusion model's stochastic sampling with edge supervision of tampered image regions mitigates the risk of incorrect predictions from overconfidence and prevents the loss of subtle boundaries that can result from overly stochastic processes. Furthermore, we propose an innovative Dual-stream Multi-scale Feature Extractor (DMFE) for extracting multi-scale features, enhancing feature representation by considering both semantic and edge conditions of the inpainted images. Extensive experiments across challenging datasets demonstrate that the InpDiffusion significantly outperforms existing state-of-the-art methods in IIL tasks, while also showcasing excellent generalization capabilities and robustness.
2501.02820
Rydberg Atomic Quantum Receivers for Multi-Target DOA Estimation
eess.SP cs.IT math.IT quant-ph
Quantum sensing technologies have experienced rapid progresses since entering the `second quantum revolution'. Among various candidates, schemes relying on Rydberg atoms exhibit compelling advantages for detecting radio frequency signals. Based on this, Rydberg atomic quantum receivers (RAQRs) have emerged as a promising solution to classical wireless communication and sensing. To harness the advantages and exploit the potential of RAQRs in wireless sensing, we investigate the realization of the direction of arrival (DOA) estimation by RAQRs. Specifically, we first conceive a Rydberg atomic quantum uniform linear array (RAQ-ULA) aided receiver for multi-target detection and propose the corresponding signal model of this sensing system. Furthermore, we propose the Rydberg atomic quantum estimation of signal parameters by designing a rotational invariance based technique termed as RAQ-ESPRIT relying on our model. The proposed algorithm solves the sensor gain mismatch problem, which is due to the presence of the RF local oscillator in the RAQ-ULA and cannot be well addressed by using the conventional ESPRIT. Lastly, we characterize our scheme through numerical simulations.
2501.02821
Targetless Intrinsics and Extrinsic Calibration of Multiple LiDARs and Cameras with IMU using Continuous-Time Estimation
cs.RO
Accurate spatiotemporal calibration is a prerequisite for multisensor fusion. However, sensors are typically asynchronous, and there is no overlap between the fields of view of cameras and LiDARs, posing challenges for intrinsic and extrinsic parameter calibration. To address this, we propose a calibration pipeline based on continuous-time and bundle adjustment (BA) capable of simultaneous intrinsic and extrinsic calibration (6 DOF transformation and time offset). We do not require overlapping fields of view or any calibration board. Firstly, we establish data associations between cameras using Structure from Motion (SFM) and perform self-calibration of camera intrinsics. Then, we establish data associations between LiDARs through adaptive voxel map construction, optimizing for extrinsic calibration within the map. Finally, by matching features between the intensity projection of LiDAR maps and camera images, we conduct joint optimization for intrinsic and extrinsic parameters. This pipeline functions in texture-rich structured environments, allowing simultaneous calibration of any number of cameras and LiDARs without the need for intricate sensor synchronization triggers. Experimental results demonstrate our method's ability to fulfill co-visibility and motion constraints between sensors without accumulating errors.
2501.02822
RDD4D: 4D Attention-Guided Road Damage Detection And Classification
cs.CV cs.AI cs.RO
Road damage detection and assessment are crucial components of infrastructure maintenance. However, current methods often struggle with detecting multiple types of road damage in a single image, particularly at varying scales. This is due to the lack of road datasets with various damage types having varying scales. To overcome this deficiency, first, we present a novel dataset called Diverse Road Damage Dataset (DRDD) for road damage detection that captures the diverse road damage types in individual images, addressing a crucial gap in existing datasets. Then, we provide our model, RDD4D, that exploits Attention4D blocks, enabling better feature refinement across multiple scales. The Attention4D module processes feature maps through an attention mechanism combining positional encoding and "Talking Head" components to capture local and global contextual information. In our comprehensive experimental analysis comparing various state-of-the-art models on our proposed, our enhanced model demonstrated superior performance in detecting large-sized road cracks with an Average Precision (AP) of 0.458 and maintained competitive performance with an overall AP of 0.445. Moreover, we also provide results on the CrackTinyNet dataset; our model achieved around a 0.21 increase in performance. The code, model weights, dataset, and our results are available on \href{https://github.com/msaqib17/Road_Damage_Detection}{https://github.com/msaqib17/Road\_Damage\_Detection}.
2501.02824
Proteomic Learning of Gamma-Aminobutyric Acid (GABA) Receptor-Mediated Anesthesia
q-bio.BM cs.LG
Anesthetics are crucial in surgical procedures and therapeutic interventions, but they come with side effects and varying levels of effectiveness, calling for novel anesthetic agents that offer more precise and controllable effects. Targeting Gamma-aminobutyric acid (GABA) receptors, the primary inhibitory receptors in the central nervous system, could enhance their inhibitory action, potentially reducing side effects while improving the potency of anesthetics. In this study, we introduce a proteomic learning of GABA receptor-mediated anesthesia based on 24 GABA receptor subtypes by considering over 4000 proteins in protein-protein interaction (PPI) networks and over 1.5 millions known binding compounds. We develop a corresponding drug-target interaction network to identify potential lead compounds for novel anesthetic design. To ensure robust proteomic learning predictions, we curated a dataset comprising 136 targets from a pool of 980 targets within the PPI networks. We employed three machine learning algorithms, integrating advanced natural language processing (NLP) models such as pretrained transformer and autoencoder embeddings. Through a comprehensive screening process, we evaluated the side effects and repurposing potential of over 180,000 drug candidates targeting the GABRA5 receptor. Additionally, we assessed the ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties of these candidates to identify those with near-optimal characteristics. This approach also involved optimizing the structures of existing anesthetics. Our work presents an innovative strategy for the development of new anesthetic drugs, optimization of anesthetic use, and deeper understanding of potential anesthesia-related side effects.
2501.02825
Randomly Sampled Language Reasoning Problems Reveal Limits of LLMs
cs.LG
Can LLMs pick up language structure from examples? Evidence in prior work seems to indicate yes, as pretrained models repeatedly demonstrate the ability to adapt to new language structures and vocabularies. However, this line of research typically considers languages that are present within common pretraining datasets, or otherwise share notable similarities with these seen languages. In contrast, in this work we attempt to measure models' language understanding capacity while circumventing the risk of dataset recall. We parameterize large families of language tasks recognized by deterministic finite automata (DFAs), and can thus sample novel language reasoning problems to fairly evaulate LLMs regardless of training data. We find that, even in the strikingly simple setting of 3-state DFAs, LLMs underperform unparameterized ngram models on both language recognition and synthesis tasks. These results suggest that LLMs struggle to match the ability of basic language models in recognizing and reasoning over languages that are sufficiently distinct from the ones they see at training time, underscoring the distinction between learning individual languages and possessing a general theory of language.
2501.02831
Universal Features Guided Zero-Shot Category-Level Object Pose Estimation
cs.CV
Object pose estimation, crucial in computer vision and robotics applications, faces challenges with the diversity of unseen categories. We propose a zero-shot method to achieve category-level 6-DOF object pose estimation, which exploits both 2D and 3D universal features of input RGB-D image to establish semantic similarity-based correspondences and can be extended to unseen categories without additional model fine-tuning. Our method begins with combining efficient 2D universal features to find sparse correspondences between intra-category objects and gets initial coarse pose. To handle the correspondence degradation of 2D universal features if the pose deviates much from the target pose, we use an iterative strategy to optimize the pose. Subsequently, to resolve pose ambiguities due to shape differences between intra-category objects, the coarse pose is refined by optimizing with dense alignment constraint of 3D universal features. Our method outperforms previous methods on the REAL275 and Wild6D benchmarks for unseen categories.
2501.02832
Samba-ASR: State-Of-The-Art Speech Recognition Leveraging Structured State-Space Models
cs.CL cs.AI cs.SD eess.AS
We propose Samba ASR,the first state of the art Automatic Speech Recognition(ASR)model leveraging the novel Mamba architecture as both encoder and decoder,built on the foundation of state space models(SSMs).Unlike transformerbased ASR models,which rely on self-attention mechanisms to capture dependencies,Samba ASR effectively models both local and global temporal dependencies using efficient statespace dynamics,achieving remarkable performance gains.By addressing the limitations of transformers,such as quadratic scaling with input length and difficulty in handling longrange dependencies,Samba ASR achieves superior accuracy and efficiency.Experimental results demonstrate that Samba ASR surpasses existing opensource transformerbased ASR models across various standard benchmarks,establishing it as the new state of theart in ASR.Extensive evaluations on the benchmark dataset show significant improvements in Word Error Rate(WER),with competitive performance even in lowresource scenarios.Furthermore,the inherent computational efficiency and parameter optimization of the Mamba architecture make Samba ASR a scalable and robust solution for diverse ASR tasks.Our contributions include the development of a new Samba ASR architecture for automatic speech recognition(ASR),demonstrating the superiority of structured statespace models(SSMs)over transformer based models for speech sequence processing.We provide a comprehensive evaluation on public benchmarks,showcasing stateoftheart(SOTA)performance,and present an indepth analysis of computational efficiency,robustness to noise,and sequence generalization.This work highlights the viability of Mamba SSMs as a transformerfree alternative for efficient and accurate ASR.By leveraging the advancements of statespace modeling,Samba ASR redefines ASR performance standards and sets a new benchmark for future research in this field.
2501.02837
Forward Once for All: Structural Parameterized Adaptation for Efficient Cloud-coordinated On-device Recommendation
cs.DC cs.AI cs.IR
In cloud-centric recommender system, regular data exchanges between user devices and cloud could potentially elevate bandwidth demands and privacy risks. On-device recommendation emerges as a viable solution by performing reranking locally to alleviate these concerns. Existing methods primarily focus on developing local adaptive parameters, while potentially neglecting the critical role of tailor-made model architecture. Insights from broader research domains suggest that varying data distributions might favor distinct architectures for better fitting. In addition, imposing a uniform model structure across heterogeneous devices may result in risking inefficacy on less capable devices or sub-optimal performance on those with sufficient capabilities. In response to these gaps, our paper introduces Forward-OFA, a novel approach for the dynamic construction of device-specific networks (both structure and parameters). Forward-OFA employs a structure controller to selectively determine whether each block needs to be assembled for a given device. However, during the training of the structure controller, these assembled heterogeneous structures are jointly optimized, where the co-adaption among blocks might encounter gradient conflicts. To mitigate this, Forward-OFA is designed to establish a structure-guided mapping of real-time behaviors to the parameters of assembled networks. Structure-related parameters and parallel components within the mapper prevent each part from receiving heterogeneous gradients from others, thus bypassing the gradient conflicts for coupled optimization. Besides, direct mapping enables Forward-OFA to achieve adaptation through only one forward pass, allowing for swift adaptation to changing interests and eliminating the requirement for on-device backpropagation. Experiments on real-world datasets demonstrate the effectiveness and efficiency of Forward-OFA.
2501.02838
Improving GenIR Systems Based on User Feedback
cs.IR
In this chapter, we discuss how to improve the GenIR systems based on user feedback. Before describing the approaches, it is necessary to be aware that the concept of "user" has been extended in the interactions with the GenIR systems. Different types of feedback information and strategies are also provided. Then the alignment techniques are highlighted in terms of objectives and methods. Following this, various ways of learning from user feedback in GenIR are presented, including continual learning, learning and ranking in the conversational context, and prompt learning. Through this comprehensive exploration, it becomes evident that innovative techniques are being proposed beyond traditional methods of utilizing user feedback, and contribute significantly to the evolution of GenIR in the new era. We also summarize some challenging topics and future directions that require further investigation.
2501.02840
Enhanced Rooftop Solar Panel Detection by Efficiently Aggregating Local Features
cs.CV cs.AI cs.LG
In this paper, we present an enhanced Convolutional Neural Network (CNN)-based rooftop solar photovoltaic (PV) panel detection approach using satellite images. We propose to use pre-trained CNN-based model to extract the local convolutional features of rooftops. These local features are then combined using the Vectors of Locally Aggregated Descriptors (VLAD) technique to obtain rooftop-level global features, which are then used to train traditional Machine Learning (ML) models to identify rooftop images that do and do not contain PV panels. On the dataset used in this study, the proposed approach achieved rooftop-PV classification scores exceeding the predefined threshold of 0.9 across all three cities for each of the feature extractor networks evaluated. Moreover, we propose a 3-phase approach to enable efficient utilization of the previously trained models on a new city or region with limited labelled data. We illustrate the effectiveness of this 3-phase approach for multi-city rooftop-PV detection task.
2501.02841
Integrating Language-Image Prior into EEG Decoding for Cross-Task Zero-Calibration RSVP-BCI
cs.HC cs.IR
Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an effective technology used for information detection by detecting Event-Related Potentials (ERPs). The current RSVP decoding methods can perform well in decoding EEG signals within a single RSVP task, but their decoding performance significantly decreases when directly applied to different RSVP tasks without calibration data from the new tasks. This limits the rapid and efficient deployment of RSVP-BCI systems for detecting different categories of targets in various scenarios. To overcome this limitation, this study aims to enhance the cross-task zero-calibration RSVP decoding performance. First, we design three distinct RSVP tasks for target image retrieval and build an open-source dataset containing EEG signals and corresponding stimulus images. Then we propose an EEG with Language-Image Prior fusion Transformer (ELIPformer) for cross-task zero-calibration RSVP decoding. Specifically, we propose a prompt encoder based on the language-image pre-trained model to extract language-image features from task-specific prompts and stimulus images as prior knowledge for enhancing EEG decoding. A cross bidirectional attention mechanism is also adopted to facilitate the effective feature fusion and alignment between the EEG and language-image features. Extensive experiments demonstrate that the proposed model achieves superior performance in cross-task zero-calibration RSVP decoding, which promotes the RSVP-BCI system from research to practical application.
2501.02842
Foundations of GenIR
cs.IR cs.LG
The chapter discusses the foundational impact of modern generative AI models on information access (IA) systems. In contrast to traditional AI, the large-scale training and superior data modeling of generative AI models enable them to produce high-quality, human-like responses, which brings brand new opportunities for the development of IA paradigms. In this chapter, we identify and introduce two of them in details, i.e., information generation and information synthesis. Information generation allows AI to create tailored content addressing user needs directly, enhancing user experience with immediate, relevant outputs. Information synthesis leverages the ability of generative AI to integrate and reorganize existing information, providing grounded responses and mitigating issues like model hallucination, which is particularly valuable in scenarios requiring precision and external knowledge. This chapter delves into the foundational aspects of generative models, including architecture, scaling, and training, and discusses their applications in multi-modal scenarios. Additionally, it examines the retrieval-augmented generation paradigm and other methods for corpus modeling and understanding, demonstrating how generative AI can enhance information access systems. It also summarizes potential challenges and fruitful directions for future studies.
2501.02843
RAHN: A Reputation Based Hourglass Network for Web Service QoS Prediction
cs.SE cs.LG
As the homogenization of Web services becomes more and more common, the difficulty of service recommendation is gradually increasing. How to predict Quality of Service (QoS) more efficiently and accurately becomes an important challenge for service recommendation. Considering the excellent role of reputation and deep learning (DL) techniques in the field of QoS prediction, we propose a reputation and DL based QoS prediction network, RAHN, which contains the Reputation Calculation Module (RCM), the Latent Feature Extraction Module (LFEM), and the QoS Prediction Hourglass Network (QPHN). RCM obtains the user reputation and the service reputation by using a clustering algorithm and a Logit model. LFEM extracts latent features from known information to form an initial latent feature vector. QPHN aggregates latent feature vectors with different scales by using Attention Mechanism, and can be stacked multiple times to obtain the final latent feature vector for prediction. We evaluate RAHN on a real QoS dataset. The experimental results show that the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of RAHN are smaller than the six baseline methods.
2501.02844
Graph-based Retrieval Augmented Generation for Dynamic Few-shot Text Classification
cs.CL cs.IR cs.LG
Text classification is a fundamental task in data mining, pivotal to various applications such as tabular understanding and recommendation. Although neural network-based models, such as CNN and BERT, have demonstrated remarkable performance in text classification, their effectiveness heavily relies on abundant labeled training data. This dependency makes these models less effective in dynamic few-shot text classification, where labeled data is scarce, and new target labels frequently appear based on application needs. Recently, large language models (LLMs) have shown promise due to their extensive pretraining and contextual understanding ability. Current approaches provide LLMs with text inputs, candidate labels, and additional side information (e.g., descriptions) to classify texts. However, their effectiveness is hindered by the increased input size and the noise introduced through side information processing. To address these limitations, we propose a graph-based online retrieval-augmented generation framework, namely GORAG, for dynamic few-shot text classification. Rather than treating each input independently, GORAG constructs and maintains a weighted graph by extracting side information across all target texts. In this graph, text keywords and labels are represented as nodes, with edges indicating the correlations between them. To model these correlations, GORAG employs an edge weighting mechanism to prioritize the importance and reliability of extracted information and dynamically retrieves relevant context using a minimum-cost spanning tree tailored for each text input. Empirical evaluations demonstrate that GORAG outperforms existing approaches by providing more comprehensive and precise contextual information.
2501.02845
HOGSA: Bimanual Hand-Object Interaction Understanding with 3D Gaussian Splatting Based Data Augmentation
cs.CV
Understanding of bimanual hand-object interaction plays an important role in robotics and virtual reality. However, due to significant occlusions between hands and object as well as the high degree-of-freedom motions, it is challenging to collect and annotate a high-quality, large-scale dataset, which prevents further improvement of bimanual hand-object interaction-related baselines. In this work, we propose a new 3D Gaussian Splatting based data augmentation framework for bimanual hand-object interaction, which is capable of augmenting existing dataset to large-scale photorealistic data with various hand-object pose and viewpoints. First, we use mesh-based 3DGS to model objects and hands, and to deal with the rendering blur problem due to multi-resolution input images used, we design a super-resolution module. Second, we extend the single hand grasping pose optimization module for the bimanual hand object to generate various poses of bimanual hand-object interaction, which can significantly expand the pose distribution of the dataset. Third, we conduct an analysis for the impact of different aspects of the proposed data augmentation on the understanding of the bimanual hand-object interaction. We perform our data augmentation on two benchmarks, H2O and Arctic, and verify that our method can improve the performance of the baselines.
2501.02850
Large Language Models for Video Surveillance Applications
cs.CV
The rapid increase in video content production has resulted in enormous data volumes, creating significant challenges for efficient analysis and resource management. To address this, robust video analysis tools are essential. This paper presents an innovative proof of concept using Generative Artificial Intelligence (GenAI) in the form of Vision Language Models to enhance the downstream video analysis process. Our tool generates customized textual summaries based on user-defined queries, providing focused insights within extensive video datasets. Unlike traditional methods that offer generic summaries or limited action recognition, our approach utilizes Vision Language Models to extract relevant information, improving analysis precision and efficiency. The proposed method produces textual summaries from extensive CCTV footage, which can then be stored for an indefinite time in a very small storage space compared to videos, allowing users to quickly navigate and verify significant events without exhaustive manual review. Qualitative evaluations result in 80% and 70% accuracy in temporal and spatial quality and consistency of the pipeline respectively.
2501.02851
Exact Matching in Correlated Networks with Node Attributes for Improved Community Recovery
cs.SI cs.IT math.IT stat.ML
We study community detection in multiple networks whose nodes and edges are jointly correlated. This setting arises naturally in applications such as social platforms, where a shared set of users may exhibit both correlated friendship patterns and correlated attributes across different platforms. Extending the classical Stochastic Block Model (SBM) and its contextual counterpart (CSBM), we introduce the correlated CSBM, which incorporates structural and attribute correlations across graphs. To build intuition, we first analyze correlated Gaussian Mixture Models, wherein only correlated node attributes are available without edges, and identify the conditions under which an estimator minimizing the distance between attributes achieves exact matching of nodes across the two databases. For correlated CSBMs, we develop a two-step procedure that first applies $k$-core matching to most nodes using edge information, then refines the matching for the remaining unmatched nodes by leveraging their attributes with a distance-based estimator. We identify the conditions under which the algorithm recovers the exact node correspondence, enabling us to merge the correlated edges and average the correlated attributes for enhanced community detection. Crucially, by aligning and combining graphs, we identify regimes in which community detection is impossible in a single graph but becomes feasible when side information from correlated graphs is incorporated. Our results illustrate how the interplay between graph matching and community recovery can boost performance, broadening the scope of multi-graph, attribute-based community detection.
2501.02855
Synthetic Fungi Datasets: A Time-Aligned Approach
cs.CV
Fungi undergo dynamic morphological transformations throughout their lifecycle, forming intricate networks as they transition from spores to mature mycelium structures. To support the study of these time-dependent processes, we present a synthetic, time-aligned image dataset that models key stages of fungal growth. This dataset systematically captures phenomena such as spore size reduction, branching dynamics, and the emergence of complex mycelium networks. The controlled generation process ensures temporal consistency, scalability, and structural alignment, addressing the limitations of real-world fungal datasets. Optimized for deep learning (DL) applications, this dataset facilitates the development of models for classifying growth stages, predicting fungal development, and analyzing morphological patterns over time. With applications spanning agriculture, medicine, and industrial mycology, this resource provides a robust foundation for automating fungal analysis, enhancing disease monitoring, and advancing fungal biology research through artificial intelligence.
2501.02857
ParetoLens: A Visual Analytics Framework for Exploring Solution Sets of Multi-objective Evolutionary Algorithms
cs.NE cs.HC cs.LG
In the domain of multi-objective optimization, evolutionary algorithms are distinguished by their capability to generate a diverse population of solutions that navigate the trade-offs inherent among competing objectives. This has catalyzed the ascension of evolutionary multi-objective optimization (EMO) as a prevalent approach. Despite the effectiveness of the EMO paradigm, the analysis of resultant solution sets presents considerable challenges. This is primarily attributed to the high-dimensional nature of the data and the constraints imposed by static visualization methods, which frequently culminate in visual clutter and impede interactive exploratory analysis. To address these challenges, this paper introduces ParetoLens, a visual analytics framework specifically tailored to enhance the inspection and exploration of solution sets derived from the multi-objective evolutionary algorithms. Utilizing a modularized, algorithm-agnostic design, ParetoLens enables a detailed inspection of solution distributions in both decision and objective spaces through a suite of interactive visual representations. This approach not only mitigates the issues associated with static visualizations but also supports a more nuanced and flexible analysis process. The usability of the framework is evaluated through case studies and expert interviews, demonstrating its potential to uncover complex patterns and facilitate a deeper understanding of multi-objective optimization solution sets. A demo website of ParetoLens is available at https://dva-lab.org/paretolens/.
2501.02858
A Novel Vision Transformer for Camera-LiDAR Fusion based Traffic Object Segmentation
cs.CV
This paper presents Camera-LiDAR Fusion Transformer (CLFT) models for traffic object segmentation, which leverage the fusion of camera and LiDAR data using vision transformers. Building on the methodology of visual transformers that exploit the self-attention mechanism, we extend segmentation capabilities with additional classification options to a diverse class of objects including cyclists, traffic signs, and pedestrians across diverse weather conditions. Despite good performance, the models face challenges under adverse conditions which underscores the need for further optimization to enhance performance in darkness and rain. In summary, the CLFT models offer a compelling solution for autonomous driving perception, advancing the state-of-the-art in multimodal fusion and object segmentation, with ongoing efforts required to address existing limitations and fully harness their potential in practical deployments.
2501.02860
Seeing the Whole in the Parts in Self-Supervised Representation Learning
cs.LG cs.CV
Recent successes in self-supervised learning (SSL) model spatial co-occurrences of visual features either by masking portions of an image or by aggressively cropping it. Here, we propose a new way to model spatial co-occurrences by aligning local representations (before pooling) with a global image representation. We present CO-SSL, a family of instance discrimination methods and show that it outperforms previous methods on several datasets, including ImageNet-1K where it achieves 71.5% of Top-1 accuracy with 100 pre-training epochs. CO-SSL is also more robust to noise corruption, internal corruption, small adversarial attacks, and large training crop sizes. Our analysis further indicates that CO-SSL learns highly redundant local representations, which offers an explanation for its robustness. Overall, our work suggests that aligning local and global representations may be a powerful principle of unsupervised category learning.
2501.02866
Constrained Multi-Modal Density Control of Linear Systems via Covariance Steering Theory
math.OC cs.SY eess.SY
In this paper, we investigate finite-horizon optimal density steering problems for discrete-time stochastic linear dynamical systems whose state probability densities can be represented as Gaussian Mixture Models (GMMs). Our goal is to compute optimal controllers that can ensure that the terminal state distribution will match the desired distribution exactly (hard-constrained version) or closely (soft-constrained version) where in the latter case we employ a Wasserstein like metric that can measure the distance between different GMMs. Our approach relies on a class of randomized control policies which allow us to reformulate the proposed density steering problems as finite-dimensional optimization problems, and in particular, linear and bilinear programs. Additionally, we explore more general density steering problems based on the approximation of general distributions by GMMs and characterize bounds for the error between the terminal distribution under our policy and the approximated GMM terminal state distribution. Finally, we demonstrate the effectiveness of our approach through non-trivial numerical experiments.
2501.02867
Diff-Lung: Diffusion-Based Texture Synthesis for Enhanced Pathological Tissue Segmentation in Lung CT Scans
eess.IV cs.CV
Accurate quantification of the extent of lung pathological patterns (fibrosis, ground-glass opacity, emphysema, consolidation) is prerequisite for diagnosis and follow-up of interstitial lung diseases. However, segmentation is challenging due to the significant class imbalance between healthy and pathological tissues. This paper addresses this issue by leveraging a diffusion model for data augmentation applied during training an AI model. Our approach generates synthetic pathological tissue patches while preserving essential shape characteristics and intricate details specific to each tissue type. This method enhances the segmentation process by increasing the occurence of underrepresented classes in the training data. We demonstrate that our diffusion-based augmentation technique improves segmentation accuracy across all pathological tissue types, particularly for the less common patterns. This advancement contributes to more reliable automated analysis of lung CT scans, potentially improving clinical decision-making and patient outcomes
2501.02869
IIMedGPT: Promoting Large Language Model Capabilities of Medical Tasks by Efficient Human Preference Alignment
cs.CL cs.AI
Recent researches of large language models(LLM), which is pre-trained on massive general-purpose corpora, have achieved breakthroughs in responding human queries. However, these methods face challenges including limited data insufficiency to support extensive pre-training and can not align responses with users' instructions. To address these issues, we introduce a medical instruction dataset, CMedINS, containing six medical instructions derived from actual medical tasks, which effectively fine-tunes LLM in conjunction with other data. Subsequently, We launch our medical model, IIMedGPT, employing an efficient preference alignment method, Direct preference Optimization(DPO). The results show that our final model outperforms existing medical models in medical dialogue.Datsets, Code and model checkpoints will be released upon acceptance.
2501.02870
Spectrum Sharing in 6G Space-Ground Integrated Networks: A Ground Protection Zone-Based Design
cs.IT cs.SY eess.SP eess.SY math.IT
Space-ground integrated network (SGIN) has been envisioned as a competitive solution for large scale and wide coverage of future wireless networks. By integrating both the non-terrestrial network (NTN) and the terrestrial network (TN), SGIN can provide high speed and omnipresent wireless network access for the users using the predefined licensed spectrums. Considering the scarcity of the spectrum resource and the low spectrum efficiency of the SGIN, we enable the NTN and TN to share the spectrum to improve overall system performance, i.e., weighted-sum area data rate (WS-ADR). However, mutual interference between NTN and TN is often inevitable and thus causes SGIN performance degradation. In this work, we consider a ground protection zone for the TN base stations, in which the NTN users are only allowed to use the NTN reserved spectrum to mitigate the NTN and TN mutual interference. We analytically derive the coverage probability and area data rate (ADR) of the typical users and study the performance under various protection zone sizes and spectrum allocation parameter settings. Simulation and numerical results demonstrate that the WS-ADR could be maximized by selecting the appropriate radius of protection zone and bandwidth allocation factor in the SGIN.
2501.02872
Two-Dimensional Unknown View Tomography from Unknown Angle Distributions
cs.CV
This study presents a technique for 2D tomography under unknown viewing angles when the distribution of the viewing angles is also unknown. Unknown view tomography (UVT) is a problem encountered in cryo-electron microscopy and in the geometric calibration of CT systems. There exists a moderate-sized literature on the 2D UVT problem, but most existing 2D UVT algorithms assume knowledge of the angle distribution which is not available usually. Our proposed methodology formulates the problem as an optimization task based on cross-validation error, to estimate the angle distribution jointly with the underlying 2D structure in an alternating fashion. We explore the algorithm's capabilities for the case of two probability distribution models: a semi-parametric mixture of von Mises densities and a probability mass function model. We evaluate our algorithm's performance under noisy projections using a PCA-based denoising technique and Graph Laplacian Tomography (GLT) driven by order statistics of the estimated distribution, to ensure near-perfect ordering, and compare our algorithm to intuitive baselines.
2501.02874
Steering Flexible Linear Objects in Planar Environments by Two Robot Hands Using Euler's Elastica Solutions
cs.RO
The manipulation of flexible objects such as cables, wires and fresh food items by robot hands forms a special challenge in robot grasp mechanics. This paper considers the steering of flexible linear objects in planar environments by two robot hands. The flexible linear object, modeled as an elastic non-stretchable rod, is manipulated by varying the gripping endpoint positions while keeping equal endpoint tangents. The flexible linear object shape has a closed form solution in terms of the grasp endpoint positions and tangents, called Euler's elastica. This paper obtains the elastica solutions under the optimal control framework, then uses the elastica solutions to obtain closed-form criteria for non self-intersection, stability and obstacle avoidance of the flexible linear object. The new tools are incorporated into a planning scheme for steering flexible linear objects in planar environments populated by sparsely spaced obstacles. The scheme is fully implemented and demonstrated with detailed examples.
2501.02880
Conditional Mutual Information Based Diffusion Posterior Sampling for Solving Inverse Problems
cs.LG stat.ML
Inverse problems are prevalent across various disciplines in science and engineering. In the field of computer vision, tasks such as inpainting, deblurring, and super-resolution are commonly formulated as inverse problems. Recently, diffusion models (DMs) have emerged as a promising approach for addressing noisy linear inverse problems, offering effective solutions without requiring additional task-specific training. Specifically, with the prior provided by DMs, one can sample from the posterior by finding the likelihood. Since the likelihood is intractable, it is often approximated in the literature. However, this approximation compromises the quality of the generated images. To overcome this limitation and improve the effectiveness of DMs in solving inverse problems, we propose an information-theoretic approach. Specifically, we maximize the conditional mutual information $\mathrm{I}(\boldsymbol{x}_0; \boldsymbol{y} | \boldsymbol{x}_t)$, where $\boldsymbol{x}_0$ represents the reconstructed signal, $\boldsymbol{y}$ is the measurement, and $\boldsymbol{x}_t$ is the intermediate signal at stage $t$. This ensures that the intermediate signals $\boldsymbol{x}_t$ are generated in a way that the final reconstructed signal $\boldsymbol{x}_0$ retains as much information as possible about the measurement $\boldsymbol{y}$. We demonstrate that this method can be seamlessly integrated with recent approaches and, once incorporated, enhances their performance both qualitatively and quantitatively.
2501.02882
PARF-Net: integrating pixel-wise adaptive receptive fields into hybrid Transformer-CNN network for medical image segmentation
cs.CV
Convolutional neural networks (CNNs) excel in local feature extraction while Transformers are superior in processing global semantic information. By leveraging the strengths of both, hybrid Transformer-CNN networks have become the major architectures in medical image segmentation tasks. However, existing hybrid methods still suffer deficient learning of local semantic features due to the fixed receptive fields of convolutions, and also fall short in effectively integrating local and long-range dependencies. To address these issues, we develop a new method PARF-Net to integrate convolutions of Pixel-wise Adaptive Receptive Fields (Conv-PARF) into hybrid Network for medical image segmentation. The Conv-PARF is introduced to cope with inter-pixel semantic differences and dynamically adjust convolutional receptive fields for each pixel, thus providing distinguishable features to disentangle the lesions with varying shapes and scales from the background. The features derived from the Conv-PARF layers are further processed using hybrid Transformer-CNN blocks under a lightweight manner, to effectively capture local and long-range dependencies, thus boosting the segmentation performance. By assessing PARF-Net on four widely used medical image datasets including MoNuSeg, GlaS, DSB2018 and multi-organ Synapse, we showcase the advantages of our method over the state-of-the-arts. For instance, PARF-Net achieves 84.27% mean Dice on the Synapse dataset, surpassing existing methods by a large margin.
2501.02885
MDP3: A Training-free Approach for List-wise Frame Selection in Video-LLMs
cs.CV cs.LG
Video large language models (Video-LLMs) have made significant progress in understanding videos. However, processing multiple frames leads to lengthy visual token sequences, presenting challenges such as the limited context length cannot accommodate the entire video, and the inclusion of irrelevant frames hinders visual perception. Hence, effective frame selection is crucial. This paper emphasizes that frame selection should follow three key principles: query relevance, list-wise diversity, and sequentiality. Existing methods, such as uniform frame sampling and query-frame matching, do not capture all of these principles. Thus, we propose Markov decision determinantal point process with dynamic programming (MDP3) for frame selection, a training-free and model-agnostic method that can be seamlessly integrated into existing Video-LLMs. Our method first estimates frame similarities conditioned on the query using a conditional Gaussian kernel within the reproducing kernel Hilbert space~(RKHS). We then apply the determinantal point process~(DPP) to the similarity matrix to capture both query relevance and list-wise diversity. To incorporate sequentiality, we segment the video and apply DPP within each segment, conditioned on the preceding segment selection, modeled as a Markov decision process~(MDP) for allocating selection sizes across segments. Theoretically, MDP3 provides a \((1 - 1/e)\)-approximate solution to the NP-hard list-wise frame selection problem with pseudo-polynomial time complexity, demonstrating its efficiency. Empirically, MDP3 significantly outperforms existing methods, verifying its effectiveness and robustness.
2501.02888
Revisiting Communication Efficiency in Multi-Agent Reinforcement Learning from the Dimensional Analysis Perspective
cs.MA
In this work, we introduce a novel perspective, i.e., dimensional analysis, to address the challenge of communication efficiency in Multi-Agent Reinforcement Learning (MARL). Our findings reveal that simply optimizing the content and timing of communication at sending end is insufficient to fully resolve communication efficiency issues. Even after applying optimized and gated messages, dimensional redundancy and confounders still persist in the integrated message embeddings at receiving end, which negatively impact communication quality and decision-making. To address these challenges, we propose Dimensional Rational Multi-Agent Communication (DRMAC), designed to mitigate both dimensional redundancy and confounders in MARL. DRMAC incorporates a redundancy-reduction regularization term to encourage the decoupling of information across dimensions within the learned representations of integrated messages. Additionally, we introduce a dimensional mask that dynamically adjusts gradient weights during training to eliminate the influence of decision-irrelevant dimensions. We evaluate DRMAC across a diverse set of multi-agent tasks, demonstrating its superior performance over existing state-of-the-art methods in complex scenarios. Furthermore, the plug-and-play nature of DRMAC's key modules highlights its generalizable performance, serving as a valuable complement rather than a replacement for existing multi-agent communication strategies.
2501.02891
Explaining Humour Style Classifications: An XAI Approach to Understanding Computational Humour Analysis
cs.CL cs.AI
Humour styles can have either a negative or a positive impact on well-being. Given the importance of these styles to mental health, significant research has been conducted on their automatic identification. However, the automated machine learning models used for this purpose are black boxes, making their prediction decisions opaque. Clarity and transparency are vital in the field of mental health. This paper presents an explainable AI (XAI) framework for understanding humour style classification, building upon previous work in computational humour analysis. Using the best-performing single model (ALI+XGBoost) from prior research, we apply comprehensive XAI techniques to analyse how linguistic, emotional, and semantic features contribute to humour style classification decisions. Our analysis reveals distinct patterns in how different humour styles are characterised and misclassified, with particular emphasis on the challenges in distinguishing affiliative humour from other styles. Through detailed examination of feature importance, error patterns, and misclassification cases, we identify key factors influencing model decisions, including emotional ambiguity, context misinterpretation, and target identification. The framework demonstrates significant utility in understanding model behaviour, achieving interpretable insights into the complex interplay of features that define different humour styles. Our findings contribute to both the theoretical understanding of computational humour analysis and practical applications in mental health, content moderation, and digital humanities research.
2501.02892
FoundPAD: Foundation Models Reloaded for Face Presentation Attack Detection
cs.CV
Although face recognition systems have seen a massive performance enhancement in recent years, they are still targeted by threats such as presentation attacks, leading to the need for generalizable presentation attack detection (PAD) algorithms. Current PAD solutions suffer from two main problems: low generalization to unknown cenarios and large training data requirements. Foundation models (FM) are pre-trained on extensive datasets, achieving remarkable results when generalizing to unseen domains and allowing for efficient task-specific adaption even when little training data are available. In this work, we recognize the potential of FMs to address common PAD problems and tackle the PAD task with an adapted FM for the first time. The FM under consideration is adapted with LoRA weights while simultaneously training a classification header. The resultant architecture, FoundPAD, is highly generalizable to unseen domains, achieving competitive results in several settings under different data availability scenarios and even when using synthetic training data. To encourage reproducibility and facilitate further research in PAD, we publicly release the implementation of FoundPAD at https://github.com/gurayozgur/FoundPAD .
2501.02893
A Volumetric Approach to Privacy of Dynamical Systems
eess.SY cs.SY
Information-theoretic metrics, such as mutual information, have been widely used to evaluate privacy leakage in dynamic systems. However, these approaches are typically limited to stochastic systems and face computational challenges. In this paper, we introduce a novel volumetric framework for analyzing privacy in systems affected by unknown but bounded noise. Our model considers a dynamic system comprising public and private states, where an observation set of the public state is released. An adversary utilizes the observed public state to infer an uncertainty set of the private state, referred to as the inference attack. We define the evolution dynamics of these inference attacks and quantify the privacy level of the private state using the volume of its uncertainty sets. For linear scalar systems, we derive an explicit formulation of the uncertainty set. For multi-dimensional linear systems, we develop an approximate computation method leveraging interval analysis. We investigate the properties of the proposed volumetric privacy measure and demonstrate that it is bounded by the information gain derived from the observation set. Furthermore, we propose an optimization approach to designing privacy filter using randomization and linear programming based on the proposed privacy measure. The effectiveness of the optimal privacy filter design is evaluated through a production-inventory case study, illustrating its robustness against the inference attack.
2501.02894
On Counting H-Intersecting Families and Graph Homomorphisms
math.CO cs.IT math.IT
This work leverages Shearer's inequalities to derive a new upper bound on the maximum cardinality of a family of graphs on a fixed number of vertices, in which every pair of graphs shares a fixed common subgraph. The derived bound is expressed in terms of the chromatic number of the shared subgraph. Additionally, Shearer's inequalities, in conjunction with properties of the Shannon entropy, are employed to establish bounds related to the enumeration of graph homomorphisms, providing further insights into the interplay between combinatorial structures and information-theoretic principles.
2501.02895
Region of Interest based Medical Image Compression
eess.IV cs.CV
The vast volume of medical image data necessitates efficient compression techniques to support remote healthcare services. This paper explores Region of Interest (ROI) coding to address the balance between compression rate and image quality. By leveraging UNET segmentation on the Brats 2020 dataset, we accurately identify tumor regions, which are critical for diagnosis. These regions are then subjected to High Efficiency Video Coding (HEVC) for compression, enhancing compression rates while preserving essential diagnostic information. This approach ensures that critical image regions maintain their quality, while non-essential areas are compressed more. Our method optimizes storage space and transmission bandwidth, meeting the demands of telemedicine and large-scale medical imaging. Through this technique, we provide a robust solution that maintains the integrity of vital data and improves the efficiency of medical image handling.
2501.02899
Finite-Sample Learning Control for LQR Over Unknown Lossy Channels
eess.SY cs.SY
This paper investigates the Linear Quadratic Regulator (LQR) problem over an unknown Bernoulli packet drop channel. The unknown packet drop probability is estimated using finite samples, then the estimated probability is used to design a formally equivalent optimal controller. If the estimation error is too large, the estimated controller cannot mean-square stabilize the system. For n-dimensional systems, the upper bound on the estimation error is provided to guarantee the stability of the closed-loop system. And we present an analytical expression for the gap between the performance of the estimated controller and the optimal performance. Next, we derive the upper bound on sample complexity for the stabilizability of the estimated controller. The tailored results with less conservatism are delivered for scalar systems and n-dimensional systems with invertible input matrix. Furthermore, a sufficient condition that does not depend on unknown channel information is provided to determine whether the estimated controller stabilizes the system with a certain probability. Finally, Numerical examples are used to demonstrate our results.
2501.02902
Sim-to-Real Transfer for Mobile Robots with Reinforcement Learning: from NVIDIA Isaac Sim to Gazebo and Real ROS 2 Robots
cs.RO cs.LG
Unprecedented agility and dexterous manipulation have been demonstrated with controllers based on deep reinforcement learning (RL), with a significant impact on legged and humanoid robots. Modern tooling and simulation platforms, such as NVIDIA Isaac Sim, have been enabling such advances. This article focuses on demonstrating the applications of Isaac in local planning and obstacle avoidance as one of the most fundamental ways in which a mobile robot interacts with its environments. Although there is extensive research on proprioception-based RL policies, the article highlights less standardized and reproducible approaches to exteroception. At the same time, the article aims to provide a base framework for end-to-end local navigation policies and how a custom robot can be trained in such simulation environment. We benchmark end-to-end policies with the state-of-the-art Nav2, navigation stack in Robot Operating System (ROS). We also cover the sim-to-real transfer process by demonstrating zero-shot transferability of policies trained in the Isaac simulator to real-world robots. This is further evidenced by the tests with different simulated robots, which show the generalization of the learned policy. Finally, the benchmarks demonstrate comparable performance to Nav2, opening the door to quick deployment of state-of-the-art end-to-end local planners for custom robot platforms, but importantly furthering the possibilities by expanding the state and action spaces or task definitions for more complex missions. Overall, with this article we introduce the most important steps, and aspects to consider, in deploying RL policies for local path planning and obstacle avoidance with Isaac Sim training, Gazebo testing, and ROS 2 for real-time inference in real robots. The code is available at https://github.com/sahars93/RL-Navigation.
2501.02905
Skillful High-Resolution Ensemble Precipitation Forecasting with an Integrated Deep Learning Framework
cs.LG cs.AI
High-resolution precipitation forecasts are crucial for providing accurate weather prediction and supporting effective responses to extreme weather events. Traditional numerical models struggle with stochastic subgrid-scale processes, while recent deep learning models often produce blurry results. To address these challenges, we propose a physics-inspired deep learning framework for high-resolution (0.05\textdegree{} $\times$ 0.05\textdegree{}) ensemble precipitation forecasting. Trained on ERA5 and CMPA high-resolution precipitation datasets, the framework integrates deterministic and probabilistic components. The deterministic model, based on a 3D SwinTransformer, captures average precipitation at mesoscale resolution and incorporates strategies to enhance performance, particularly for moderate to heavy rainfall. The probabilistic model employs conditional diffusion in latent space to account for uncertainties in residual precipitation at convective scales. During inference, ensemble members are generated by repeatedly sampling latent variables, enabling the model to represent precipitation uncertainty. Our model significantly enhances spatial resolution and forecast accuracy. Rank histogram shows that the ensemble system is reliable and unbiased. In a case study of heavy precipitation in southern China, the model outputs align more closely with observed precipitation distributions than ERA5, demonstrating superior capability in capturing extreme precipitation events. Additionally, 5-day real-time forecasts show good performance in terms of CSI scores.
2501.02906
Domain-Agnostic Co-Evolution of Generalizable Parallel Algorithm Portfolios
cs.NE
Generalization is the core objective when training optimizers from data. However, limited training instances often constrain the generalization capability of the trained optimizers. Co-evolutionary approaches address this challenge by simultaneously evolving a parallel algorithm portfolio (PAP) and an instance population to eventually obtain PAPs with good generalization. Yet, when applied to a specific problem class, these approaches have a major limitation. They require practitioners to provide instance generators specially tailored to the problem class, which is often non-trivial to design. This work proposes a general-purpose, off-the-shelf PAP construction approach, named domain-agnostic co-evolution of parameterized search (DACE), for binary optimization problems where decision variables take values of 0 or 1. The key innovation of DACE lies in its neural network-based domain-agnostic instance representation and generation mechanism that delimitates the need for domain-specific instance generators. The strong generality of DACE is validated across three real-world binary optimization problems: the complementary influence maximization problem (CIMP), the compiler arguments optimization problem (CAOP), and the contamination control problem (CCP). Given only a small set of training instances from these classes, DACE, without requiring any domain knowledge, constructs PAPs with better generalization performance than existing approaches on all three classes, despite their use of domain-specific instance generators.
2501.02909
Comprehensive Pathological Image Segmentation via Teacher Aggregation for Tumor Microenvironment Analysis
cs.CV
The tumor microenvironment (TME) plays a crucial role in cancer progression and treatment response, yet current methods for its comprehensive analysis in H&E-stained tissue slides face significant limitations in the diversity of tissue cell types and accuracy. Here, we present PAGET (Pathological image segmentation via AGgrEgated Teachers), a new knowledge distillation approach that integrates multiple segmentation models while considering the hierarchical nature of cell types in the TME. By leveraging a unique dataset created through immunohistochemical restaining techniques and existing segmentation models, PAGET enables simultaneous identification and classification of 14 key TME components. We demonstrate PAGET's ability to perform rapid, comprehensive TME segmentation across various tissue types and medical institutions, advancing the quantitative analysis of tumor microenvironments. This method represents a significant step forward in enhancing our understanding of cancer biology and supporting precise clinical decision-making from large-scale histopathology images.
2501.02913
Pointmap-Conditioned Diffusion for Consistent Novel View Synthesis
cs.CV
In this paper, we present PointmapDiffusion, a novel framework for single-image novel view synthesis (NVS) that utilizes pre-trained 2D diffusion models. Our method is the first to leverage pointmaps (i.e. rasterized 3D scene coordinates) as a conditioning signal, capturing geometric prior from the reference images to guide the diffusion process. By embedding reference attention blocks and a ControlNet for pointmap features, our model balances between generative capability and geometric consistency, enabling accurate view synthesis across varying viewpoints. Extensive experiments on diverse real-world datasets demonstrate that PointmapDiffusion achieves high-quality, multi-view consistent results with significantly fewer trainable parameters compared to other baselines for single-image NVS tasks.
2501.02916
Spiking monocular event based 6D pose estimation for space application
cs.CV cs.LG
With the growing interest in on On-orbit servicing (OOS) and Active Debris Removal (ADR) missions, spacecraft poses estimation algorithms are being developed using deep learning to improve the precision of this complex task and find the most efficient solution. With the advances of bio-inspired low-power solutions, such a spiking neural networks and event-based processing and cameras, and their recent work for space applications, we propose to investigate the feasibility of a fully event-based solution to improve event-based pose estimation for spacecraft. In this paper, we address the first event-based dataset SEENIC with real event frames captured by an event-based camera on a testbed. We show the methods and results of the first event-based solution for this use case, where our small spiking end-to-end network (S2E2) solution achieves interesting results over 21cm position error and 14degree rotation error, which is the first step towards fully event-based processing for embedded spacecraft pose estimation.
2501.02917
On Achievable Rates Over Noisy Nanopore Channels
cs.IT math.IT
In this paper, we consider a recent channel model of a nanopore sequencer proposed by McBain, Viterbo, and Saunderson (2024), termed the noisy nanopore channel (NNC). In essence, an NNC is a noisy duplication channel, whose input source has a specific Markov structure. We present bounds on the channel capacity of selected NNCs, via simple information-theoretic inequalities. In particular, we provide a (tight) lower bound on the capacity of the noiseless NCC and demonstrate that for an NNC with erasure noise, the capacity approaches $1$ for nanopore memories that scale roughly logarithmically in the length of the input sequence.
2501.02921
Unsupervised Tomato Split Anomaly Detection using Hyperspectral Imaging and Variational Autoencoders
cs.CV cs.AI
Tomato anomalies/damages pose a significant challenge in greenhouse farming. While this method of cultivation benefits from efficient resource utilization, anomalies can significantly degrade the quality of farm produce. A common anomaly associated with tomatoes is splitting, characterized by the development of cracks on the tomato skin, which degrades its quality. Detecting this type of anomaly is challenging due to dynamic variations in appearance and sizes, compounded by dataset scarcity. We address this problem in an unsupervised manner by utilizing a tailored variational autoencoder (VAE) with hyperspectral input. Preliminary analysis of the dataset enabled us to select the optimal range of wavelengths for detecting this anomaly. Our findings indicate that the 530nm - 550nm range is suitable for identifying tomato dry splits. The analysis on reconstruction loss allow us to not only detect the anomalies but also to some degree estimate the anomalous regions.
2501.02922
Label-free Concept Based Multiple Instance Learning for Gigapixel Histopathology
cs.CV cs.AI
Multiple Instance Learning (MIL) methods allow for gigapixel Whole-Slide Image (WSI) analysis with only slide-level annotations. Interpretability is crucial for safely deploying such algorithms in high-stakes medical domains. Traditional MIL methods offer explanations by highlighting salient regions. However, such spatial heatmaps provide limited insights for end users. To address this, we propose a novel inherently interpretable WSI-classification approach that uses human-understandable pathology concepts to generate explanations. Our proposed Concept MIL model leverages recent advances in vision-language models to directly predict pathology concepts based on image features. The model's predictions are obtained through a linear combination of the concepts identified on the top-K patches of a WSI, enabling inherent explanations by tracing each concept's influence on the prediction. In contrast to traditional concept-based interpretable models, our approach eliminates the need for costly human annotations by leveraging the vision-language model. We validate our method on two widely used pathology datasets: Camelyon16 and PANDA. On both datasets, Concept MIL achieves AUC and accuracy scores over 0.9, putting it on par with state-of-the-art models. We further find that 87.1\% (Camelyon16) and 85.3\% (PANDA) of the top 20 patches fall within the tumor region. A user study shows that the concepts identified by our model align with the concepts used by pathologists, making it a promising strategy for human-interpretable WSI classification.
2501.02926
Offline-to-online hyperparameter transfer for stochastic bandits
cs.LG
Classic algorithms for stochastic bandits typically use hyperparameters that govern their critical properties such as the trade-off between exploration and exploitation. Tuning these hyperparameters is a problem of great practical significance. However, this is a challenging problem and in certain cases is information theoretically impossible. To address this challenge, we consider a practically relevant transfer learning setting where one has access to offline data collected from several bandit problems (tasks) coming from an unknown distribution over the tasks. Our aim is to use this offline data to set the hyperparameters for a new task drawn from the unknown distribution. We provide bounds on the inter-task (number of tasks) and intra-task (number of arm pulls for each task) sample complexity for learning near-optimal hyperparameters on unseen tasks drawn from the distribution. Our results apply to several classic algorithms, including tuning the exploration parameters in UCB and LinUCB and the noise parameter in GP-UCB. Our experiments indicate the significance and effectiveness of the transfer of hyperparameters from offline problems in online learning with stochastic bandit feedback.
2501.02928
Deep Generative Model-Aided Power System Dynamic State Estimation and Reconstruction with Unknown Control Inputs or Data Distributions
eess.SY cs.SY
Fast and robust dynamic state estimation (DSE) is essential for accurately capturing the internal dynamic processes of power systems, and it serves as the foundation for reliably implementing real-time dynamic modeling, monitoring, and control applications. Nonetheless, on one hand, traditional DSE methods based on Kalman filtering or particle filtering have high accuracy requirements for system parameters, control inputs, phasor measurement unit (PMU) data, and centralized DSE communication. Consequently, these methods often face accuracy bottlenecks when dealing with structural or system process errors, unknown control vectors, PMU anomalies, and communication contingencies. On the other hand, deep learning-aided DSE, while parameter-free, often suffers from generalization issues under unforeseen operating conditions. To address these challenges, this paper proposes an effective approach that leverages deep generative models from AI-generated content (AIGC) to assist DSE. The proposed approach employs an encoder-decoder architecture to estimate unknown control input variables, a robust encoder to mitigate the impact of bad PMU data, and latent diffusion model to address communication issues in centralized DSE. Additionally, a lightweight adaptor is designed to quickly adjust the latent vector distribution. Extensive experimental results on the IEEE 39-bus system and the NPCC 140-bus system demonstrate the effectiveness and superiority of the proposed method in addressing DSE modeling imperfection, measurement uncertainties, communication contingencies, and unknown distribution challenges, while also proving its ability to reduce data storage and communication resource requirements.
2501.02931
Self-Attention as a Parametric Endofunctor: A Categorical Framework for Transformer Architectures
cs.LG
Self-attention mechanisms have revolutionised deep learning architectures, yet their core mathematical structures remain incompletely understood. In this work, we develop a category-theoretic framework focusing on the linear components of self-attention. Specifically, we show that the query, key, and value maps naturally define a parametric 1-morphism in the 2-category $\mathbf{Para(Vect)}$. On the underlying 1-category $\mathbf{Vect}$, these maps induce an endofunctor whose iterated composition precisely models multi-layer attention. We further prove that stacking multiple self-attention layers corresponds to constructing the free monad on this endofunctor. For positional encodings, we demonstrate that strictly additive embeddings correspond to monoid actions in an affine sense, while standard sinusoidal encodings, though not additive, retain a universal property among injective (faithful) position-preserving maps. We also establish that the linear portions of self-attention exhibit natural equivariance to permutations of input tokens, and show how the "circuits" identified in mechanistic interpretability can be interpreted as compositions of parametric 1-morphisms. This categorical perspective unifies geometric, algebraic, and interpretability-based approaches to transformer analysis, making explicit the underlying structures of attention. We restrict to linear maps throughout, deferring the treatment of nonlinearities such as softmax and layer normalisation, which require more advanced categorical constructions. Our results build on and extend recent work on category-theoretic foundations for deep learning, offering deeper insights into the algebraic structure of attention mechanisms.
2501.02932
Predicting band gap from chemical composition: A simple learned model for a material property with atypical statistics
cond-mat.mtrl-sci cs.LG physics.chem-ph
In solid-state materials science, substantial efforts have been devoted to the calculation and modeling of the electronic band gap. While a wide range of ab initio methods and machine learning algorithms have been created that can predict this quantity, the development of new computational approaches for studying the band gap remains an active area of research. Here we introduce a simple machine learning model for predicting the band gap using only the chemical composition of the crystalline material. To motivate the form of the model, we first analyze the empirical distribution of the band gap, which sheds new light on its atypical statistics. Specifically, our analysis enables us to frame band gap prediction as a task of modeling a mixed random variable, and we design our model accordingly. Our model formulation incorporates thematic ideas from chemical heuristic models for other material properties in a manner that is suited towards the band gap modeling task. The model has exactly one parameter corresponding to each element, which is fit using data. To predict the band gap for a given material, the model computes a weighted average of the parameters associated with its constituent elements and then takes the maximum of this quantity and zero. The model provides heuristic chemical interpretability by intuitively capturing the associations between the band gap and individual chemical elements.
2501.02934
A Bayesian Approach for Discovering Time- Delayed Differential Equation from Data
stat.ML cs.LG physics.comp-ph
Time-delayed differential equations (TDDEs) are widely used to model complex dynamic systems where future states depend on past states with a delay. However, inferring the underlying TDDEs from observed data remains a challenging problem due to the inherent nonlinearity, uncertainty, and noise in real-world systems. Conventional equation discovery methods often exhibit limitations when dealing with large time delays, relying on deterministic techniques or optimization-based approaches that may struggle with scalability and robustness. In this paper, we present BayTiDe - Bayesian Approach for Discovering Time-Delayed Differential Equations from Data, that is capable of identifying arbitrarily large values of time delay to an accuracy that is directly proportional to the resolution of the data input to it. BayTiDe leverages Bayesian inference combined with a sparsity-promoting discontinuous spike-and-slab prior to accurately identify time-delayed differential equations. The approach accommodates arbitrarily large time delays with accuracy proportional to the input data resolution, while efficiently narrowing the search space to achieve significant computational savings. We demonstrate the efficiency and robustness of BayTiDe through a range of numerical examples, validating its ability to recover delayed differential equations from noisy data.
2501.02937
4D-CS: Exploiting Cluster Prior for 4D Spatio-Temporal LiDAR Semantic Segmentation
cs.CV
Semantic segmentation of LiDAR points has significant value for autonomous driving and mobile robot systems. Most approaches explore spatio-temporal information of multi-scan to identify the semantic classes and motion states for each point. However, these methods often overlook the segmentation consistency in space and time, which may result in point clouds within the same object being predicted as different categories. To handle this issue, our core idea is to generate cluster labels across multiple frames that can reflect the complete spatial structure and temporal information of objects. These labels serve as explicit guidance for our dual-branch network, 4D-CS, which integrates point-based and cluster-based branches to enable more consistent segmentation. Specifically, in the point-based branch, we leverage historical knowledge to enrich the current feature through temporal fusion on multiple views. In the cluster-based branch, we propose a new strategy to produce cluster labels of foreground objects and apply them to gather point-wise information to derive cluster features. We then merge neighboring clusters across multiple scans to restore missing features due to occlusion. Finally, in the point-cluster fusion stage, we adaptively fuse the information from the two branches to optimize segmentation results. Extensive experiments confirm the effectiveness of the proposed method, and we achieve state-of-the-art results on the multi-scan semantic and moving object segmentation on SemanticKITTI and nuScenes datasets. The code will be available at https://github.com/NEU-REAL/4D-CS.git.