arXiv subjects

Yang Lu

Publications and source records attributed to Yang Lu.

At least 91 records · Page 5Linked to original sources

arXiv2025

From Perception to Cognition: A Survey of Vision-Language Interactive Reasoning in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) strive to achieve a profound, human-like understanding of and interaction with the physical world, but often exhibit a shallow and incoherent integration when acquiring information (Perception) and conducting reasoning (Cognition). This disconnect leads to a spectrum of reasoning failures, with hallucination being the most prominent. Collectively, these issues expose a fundamental challenge: the ability to process pixels does not yet confer the ability to construct a coherent, credible internal world model. To systematically dissect and address this challenge, this survey introduces a novel and unified analytical framework: ``From Perception to Cognition." We deconstruct the complex process of vision-language interactive understanding into two interdependent layers: Perception, the foundational ability to accurately extract visual information and achieve fine-grained alignment with textual instructions; and Cognition, the higher-order capability for proactive, multi-step, goal-oriented reasoning built upon this perceptual foundation, the core of which is the formation of a dynamic observe-think-verify reasoning loop. Guided by this framework, this paper systematically analyzes the key bottlenecks of current MLLMs at both layers. It surveys the landscape of cutting-edge methods designed to address these challenges, spanning from techniques that enhance low-level visual representations to those that improve high-level reasoning paradigms. Furthermore, we review critical benchmarks and delineate future research directions. This survey aims to provide the research community with a clear, structured perspective for understanding the intrinsic limitations of current MLLMs and to illuminate the path toward building next-generation models capable of deep reasoning and a genuine understanding of the world.

cs.AI↗

arXiv2025

Evaluating the Quality of Randomness and Entropy in Tasks Supported by Large Language Models

The rapid advancement of large language model (LLM) technology has led to diverse applications, many of which inherently require randomness, such as stochastic decision-making, gaming, scheduling, AI agents, and cryptography-related tasks. However, the capabilities of LLMs in handling randomness, particularly in generating and utilizing random numbers effectively, remain unclear. This paper investigates the capacity of LLMs for handling tasks that involve randomness through a series of experiments. We designed a set of experiments that consider various factors that can influence an LLM's performance in tasks involving randomness, such as accessibility to external tools, types of tasks, model states (fresh vs. non-fresh), and prompting strategies. The experiments cover a range of tasks, including generating random numbers, generating random strings such as passwords, shuffling items, and evaluating the quality of randomness using entropy and the NIST randomness test-suite. Our findings reveal that while LLMs can generate outputs that exhibit some degree of randomness, their performance is inconsistent and often deviates significantly from the expected behavior. The analysis of the experimental results highlights key limitations and areas where improvement is needed for the LLMs to effectively handle tasks involving randomness

cs.AI↗

arXiv2025

PRO-VPT: Distribution-Adaptive Visual Prompt Tuning via Prompt Relocation

Visual prompt tuning (VPT), i.e., fine-tuning some lightweight prompt tokens, provides an efficient and effective approach for adapting pre-trained models to various downstream tasks. However, most prior art indiscriminately uses a fixed prompt distribution across different tasks, neglecting the importance of each block varying depending on the task. In this paper, we introduce adaptive distribution optimization (ADO) by tackling two key questions: (1) How to appropriately and formally define ADO, and (2) How to design an adaptive distribution strategy guided by this definition? Through empirical analysis, we first confirm that properly adjusting the distribution significantly improves VPT performance, and further uncover a key insight that a nested relationship exists between ADO and VPT. Based on these findings, we propose a new VPT framework, termed PRO-VPT (iterative Prompt RelOcation-based VPT), which adaptively adjusts the distribution built upon a nested optimization formulation. Specifically, we develop a prompt relocation strategy derived from this formulation, comprising two steps: pruning idle prompts from prompt-saturated blocks, followed by allocating these prompts to the most prompt-needed blocks. By iteratively performing prompt relocation and VPT, our proposal can adaptively learn the optimal prompt distribution in a nested optimization-based manner, thereby unlocking the full potential of VPT. Extensive experiments demonstrate that our proposal significantly outperforms advanced VPT methods, e.g., PRO-VPT surpasses VPT by 1.6 pp and 2.0 pp average accuracy, leading prompt-based methods to state-of-the-art performance on VTAB-1k and FGVC benchmarks. The code is available at https://github.com/ckshang/PRO-VPT.

cs.CV↗

arXiv2025

Transparent Reference-free Automated Evaluation of Open-Ended User Survey Responses

Open-ended survey responses provide valuable insights in marketing research, but low-quality responses not only burden researchers with manual filtering but also risk leading to misleading conclusions, underscoring the need for effective evaluation. Existing automatic evaluation methods target LLM-generated text and inadequately assess human-written responses with their distinct characteristics. To address such characteristics, we propose a two-stage evaluation framework specifically designed for human survey responses. First, gibberish filtering removes nonsensical responses. Then, three dimensions-effort, relevance, and completeness-are evaluated using LLM capabilities, grounded in empirical analysis of real-world survey data. Validation on English and Korean datasets shows that our framework not only outperforms existing metrics but also demonstrates high practical applicability for real-world applications such as response quality prediction and response rejection, showing strong correlations with expert assessment.

cs.CL↗

arXiv2025

You Are Your Own Best Teacher: Achieving Centralized-level Performance in Federated Learning under Heterogeneous and Long-tailed Data

Data heterogeneity, stemming from local non-IID data and global long-tailed distributions, is a major challenge in federated learning (FL), leading to significant performance gaps compared to centralized learning. Previous research found that poor representations and biased classifiers are the main problems and proposed neural-collapse-inspired synthetic simplex ETF to help representations be closer to neural collapse optima. However, we find that the neural-collapse-inspired methods are not strong enough to reach neural collapse and still have huge gaps to centralized training. In this paper, we rethink this issue from a self-bootstrap perspective and propose FedYoYo (You Are Your Own Best Teacher), introducing Augmented Self-bootstrap Distillation (ASD) to improve representation learning by distilling knowledge between weakly and strongly augmented local samples, without needing extra datasets or models. We further introduce Distribution-aware Logit Adjustment (DLA) to balance the self-bootstrap process and correct biased feature representations. FedYoYo nearly eliminates the performance gap, achieving centralized-level performance even under mixed heterogeneity. It enhances local representation learning, reducing model drift and improving convergence, with feature prototypes closer to neural collapse optimality. Extensive experiments show FedYoYo achieves state-of-the-art results, even surpassing centralized logit adjustment methods by 5.4\% under global long-tailed settings.

cs.LG↗

arXiv2025

Vector Field-Guided Learning Predictive Control for Motion Planning of Mobile Robots with Uncertain Dynamics

In obstacle-dense scenarios, providing safe guidance for mobile robots is critical to improve the safe maneuvering capability. However, the guidance provided by standard guiding vector fields (GVFs) may limit the motion capability due to the improper curvature of the integral curve when traversing obstacles. On the other hand, robotic system dynamics are often time-varying, uncertain, and even unknown during the motion planning process. Therefore, many existing kinodynamic motion planning methods could not achieve satisfactory reliability in guaranteeing safety. To address these challenges, we propose a two-level Vector Field-guided Learning Predictive Control (VF-LPC) approach that improves safe maneuverability. The first level, the guiding level, generates safe desired trajectories using the designed kinodynamic GVF, enabling safe motion in obstacle-dense environments. The second level, the Integrated Motion Planning and Control (IMPC) level, first uses a deep Koopman operator to learn a nominal dynamics model offline and then updates the model uncertainties online using sparse Gaussian processes (GPs). The learned dynamics and a game-based safe barrier function are then incorporated into the LPC framework to generate near-optimal planning solutions. Extensive simulations and real-world experiments were conducted on quadrotor unmanned aerial vehicles and unmanned ground vehicles, demonstrating that VF-LPC enables robots to maneuver safely.

cs.RO↗

arXiv2025

DPSformer: A long-tail-aware model for improving heavy rainfall prediction

Accurate and timely forecasting of heavy rainfall remains a critical challenge for modern society. Precipitation exhibits a highly imbalanced distribution: most observations record no or light rain, while heavy rainfall events are rare. Such an imbalanced distribution obstructs deep learning models from effectively predicting heavy rainfall events. To address this challenge, we treat rainfall forecasting explicitly as a long-tailed learning problem, identifying the insufficient representation of heavy rainfall events as the primary barrier to forecasting accuracy. Therefore, we introduce DPSformer, a long-tail-aware model that enriches representation of heavy rainfall events through a high-resolution branch. For heavy rainfall events $ \geq $ 50 mm/6 h, DPSformer lifts the Critical Success Index (CSI) of a baseline Numerical Weather Prediction (NWP) model from 0.012 to 0.067. For the top 1% coverage of heavy rainfall events, its Fraction Skill Score (FSS) exceeds 0.45, surpassing existing methods. Our work establishes an effective long-tailed paradigm for heavy rainfall prediction, offering a practical tool to enhance early warning systems and mitigate the societal impacts of extreme weather events.

cs.LG↗

arXiv2025

Large Language Models for Next-Generation Wireless Network Management: A Survey and Tutorial

The rapid advancement toward sixth-generation (6G) wireless networks has significantly intensified the complexity and scale of optimization problems, including resource allocation and trajectory design, often formulated as combinatorial problems in large discrete decision spaces. However, traditional optimization methods, such as heuristics and deep reinforcement learning (DRL), struggle to meet the demanding requirements of real-time adaptability, scalability, and dynamic handling of user intents in increasingly heterogeneous and resource-constrained network environments. Large language models (LLMs) present a transformative paradigm by enabling natural language-driven problem formulation, context-aware reasoning, and adaptive solution refinement through advanced semantic understanding and structured reasoning capabilities. This paper provides a systematic and comprehensive survey of LLM-enabled optimization frameworks tailored for wireless networks. We first introduce foundational design concepts and distinguish LLM-enabled methods from conventional optimization paradigms. Subsequently, we critically analyze key enabling methodologies, including natural language modeling, solver collaboration, and solution verification processes. Moreover, we explore representative case studies to demonstrate LLMs' transformative potential in practical scenarios such as optimization formulation, low-altitude economy networking, and intent networking. Finally, we discuss current research challenges, examine prominent open-source frameworks and datasets, and identify promising future directions to facilitate robust, scalable, and trustworthy LLM-enabled optimization solutions for next-generation wireless networks.

cs.NI↗

arXiv2025

Novel Category Discovery with X-Agent Attention for Open-Vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation (OVSS) conducts pixel-level classification via text-driven alignment, where the domain discrepancy between base category training and open-vocabulary inference poses challenges in discriminative modeling of latent unseen category. To address this challenge, existing vision-language model (VLM)-based approaches demonstrate commendable performance through pre-trained multi-modal representations. However, the fundamental mechanisms of latent semantic comprehension remain underexplored, making the bottleneck for OVSS. In this work, we initiate a probing experiment to explore distribution patterns and dynamics of latent semantics in VLMs under inductive learning paradigms. Building on these insights, we propose X-Agent, an innovative OVSS framework employing latent semantic-aware ``agent'' to orchestrate cross-modal attention mechanisms, simultaneously optimizing latent semantic dynamic and amplifying its perceptibility. Extensive benchmark evaluations demonstrate that X-Agent achieves state-of-the-art performance while effectively enhancing the latent semantic saliency.

cs.CV↗

arXiv2025

Categorical Data Clustering via Value Order Estimated Distance Metric Learning

Clustering is a popular machine learning technique for data mining that can process and analyze datasets to automatically reveal sample distribution patterns. Since the ubiquitous categorical data naturally lack a well-defined metric space such as the Euclidean distance space of numerical data, the distribution of categorical data is usually under-represented, and thus valuable information can be easily twisted in clustering. This paper, therefore, introduces a novel order distance metric learning approach to intuitively represent categorical attribute values by learning their optimal order relationship and quantifying their distance in a line similar to that of the numerical attributes. Since subjectively created qualitative categorical values involve ambiguity and fuzziness, the order distance metric is learned in the context of clustering. Accordingly, a new joint learning paradigm is developed to alternatively perform clustering and order distance metric learning with low time complexity and a guarantee of convergence. Due to the clustering-friendly order learning mechanism and the homogeneous ordinal nature of the order distance and Euclidean distance, the proposed method achieves superior clustering accuracy on categorical and mixed datasets. More importantly, the learned order distance metric greatly reduces the difficulty of understanding and managing the non-intuitive categorical data. Experiments with ablation studies, significance tests, case studies, etc., have validated the efficacy of the proposed method. The source code is available at https://github.com/DAJ0612/OCL_Source_Code.

cs.LG↗

arXiv2025

T-Stars-Poster: A Framework for Product-Centric Advertising Image Design

Creating advertising images is often a labor-intensive and time-consuming process. Can we automatically generate such images using basic product information like a product foreground image, taglines, and a target size? Existing methods mainly focus on parts of the problem and lack a comprehensive solution. To bridge this gap, we propose a novel product-centric framework for advertising image design called T-Stars-Poster. It consists of four sequential stages to highlight product foregrounds and taglines while achieving overall image aesthetics: prompt generation, layout generation, background image generation, and graphics rendering. Different expert models are designed and trained for the first three stages: First, a visual language model (VLM) generates background prompts that match the products. Next, a VLM-based layout generation model arranges the placement of product foregrounds, graphic elements (taglines and decorative underlays), and various nongraphic elements (objects from the background prompt). Following this, an SDXL-based model can simultaneously accept prompts, layouts, and foreground controls to generate images. To support T-Stars-Poster, we create two corresponding datasets with over 50,000 labeled images. Extensive experiments and online A/B tests demonstrate that T-Stars-Poster can produce more visually appealing advertising images.

cs.CV↗

arXiv2025

Examining the Association between Estimated Prevalence and Diagnostic Test Accuracy using Directed Acyclic Graphs

There have been reports of correlation between estimates of prevalence and test accuracy across studies included in diagnostic meta-analyses. It has been hypothesized that this unexpected association arises because of certain biases commonly found in diagnostic accuracy studies. A theoretical explanation has not been studied systematically. In this work, we introduce directed acyclic graphs to illustrate common structures of bias in diagnostic test accuracy studies and to define the resulting data-generating mechanism behind a diagnostic meta-analysis. Using simulation studies, we examine how these common biases can produce a correlation between estimates of prevalence and index test accuracy and what factors influence its magnitude and direction. We found that an association arises either in the absence of a perfect reference test or in the presence of a covariate that simultaneously causes spectrum effect and is associated with the prevalence (confounding). We also show that the association between prevalence and accuracy can be removed by appropriate statistical methods. In the risk of bias evaluation in diagnostic meta-analyses, an observed association between estimates of prevalence and accuracy should be explored to understand its source and to adjust for latent or observed variables if possible.

stat.ME↗

arXiv2025

Agentic Graph Neural Networks for Wireless Communications and Networking Towards Edge General Intelligence: A Survey

The rapid advancement of communication technologies has driven the evolution of communication networks towards both high-dimensional resource utilization and multifunctional integration. This evolving complexity poses significant challenges in designing communication networks to satisfy the growing quality-of-service and time sensitivity of mobile applications in dynamic environments. Graph neural networks (GNNs) have emerged as fundamental deep learning (DL) models for complex communication networks. GNNs not only augment the extraction of features over network topologies but also enhance scalability and facilitate distributed computation. However, most existing GNNs follow a traditional passive learning framework, which may fail to meet the needs of increasingly diverse wireless systems. This survey proposes the employment of agentic artificial intelligence (AI) to organize and integrate GNNs, enabling scenario- and task-aware implementation towards edge general intelligence. To comprehend the full capability of GNNs, we holistically review recent applications of GNNs in wireless communications and networking. Specifically, we focus on the alignment between graph representations and network topologies, and between neural architectures and wireless tasks. We first provide an overview of GNNs based on prominent neural architectures, followed by the concept of agentic GNNs. Then, we summarize and compare GNN applications for conventional systems and emerging technologies, including physical, MAC, and network layer designs, integrated sensing and communication (ISAC), reconfigurable intelligent surface (RIS) and cell-free network architecture. We further propose a large language model (LLM) framework as an intelligent question-answering agent, leveraging this survey as a local knowledge base to enable GNN-related responses tailored to wireless communication research.

eess.SP↗

arXiv2025

Mamba for Wireless Communications and Networking: Principles and Opportunities

Mamba has emerged as a powerful model for efficiently addressing tasks involving temporal and spatial data. Regarding the escalating heterogeneity and dynamics in wireless networks, Mamba holds the potential to revolutionize wireless communication and networking designs by balancing the trade-off between computational efficiency and effectiveness. This article presents a comprehensive overview of Mamba' applications in wireless systems. Specifically, we first analyze the potentials of Mamba for wireless signal processing tasks from the perspectives of long-range dependency modeling and spatial feature extraction. Then we propose two application frameworks for Mamba in wireless communications, i.e., replacement of traditional algorithms, and enabler of novel paradigms. Guided by the two frameworks, we conduct case studies on intelligent resource allocation and joint source and channel decoding to demonstrate Mamba's improvements in both feature enhancement and computational efficiency. Finally, we highlight critical challenges and outline potential research directions for Mamba in wireless communications and networking.

cs.NI↗

arXiv2025

Anyone Can Jailbreak: Prompt-Based Attacks on LLMs and T2Is

Despite significant advancements in alignment and content moderation, large language models (LLMs) and text-to-image (T2I) systems remain vulnerable to prompt-based attacks known as jailbreaks. Unlike traditional adversarial examples requiring expert knowledge, many of today's jailbreaks are low-effort, high-impact crafted by everyday users with nothing more than cleverly worded prompts. This paper presents a systems-style investigation into how non-experts reliably circumvent safety mechanisms through techniques such as multi-turn narrative escalation, lexical camouflage, implication chaining, fictional impersonation, and subtle semantic edits. We propose a unified taxonomy of prompt-level jailbreak strategies spanning both text-output and T2I models, grounded in empirical case studies across popular APIs. Our analysis reveals that every stage of the moderation pipeline, from input filtering to output validation, can be bypassed with accessible strategies. We conclude by highlighting the urgent need for context-aware defenses that reflect the ease with which these jailbreaks can be reproduced in real-world settings.

cs.CV↗

arXiv2025

Covert Communications in MEC-Based Networked ISAC Systems Towards Low-Altitude Economy

Low-altitude economy (LAE) is an emerging business model, which heavily relies on integrated sensing and communications (ISAC), mobile edge computing (MEC), and covert communications. This paper investigates the convert transmission design in MEC-based networked ISAC systems towards LAE, where an MEC server coordinates multiple access points to simultaneously receive computation tasks from multiple unmanned aerial vehicles (UAVs), locate a target in a sensing area, and maintain UAVs' covert transmission against multiple wardens. We first derive closed-form expressions for the detection error probability (DEP) at wardens. Then, we formulate a total energy consumption minimization problem by optimizing communication, sensing, and computation resources as well as UAV trajectories, subject to the requirements on quality of MEC services, DEP, and radar signal-to-interference-and-noise ratio, and the causality of UAV trajectories. An alternating optimization based algorithm is proposed to handle the considered problem, which decomposes it into two subproblems: joint optimization of communication, sensing, and computation resources, and UAV trajectory optimization. The former is addressed by a successive convex approximation based algorithm, while the latter is solved via a trust-region based algorithm. Simulations validate the effectiveness of the proposed algorithm compared with various benchmarks, and reveal the trade-offs among communication, sensing, and computation in LAE systems.

cs.IT↗

arXiv2025

From General Reasoning to Domain Expertise: Uncovering the Limits of Generalization in Large Language Models

Recent advancements in Large Language Models (LLMs) have demonstrated remarkable capabilities in various domains. However, effective decision-making relies heavily on strong reasoning abilities. Reasoning is the foundation for decision-making, providing the analytical and logical framework to make sound choices. Reasoning involves analyzing information, drawing inferences, and reaching conclusions based on logic or evidence. Decision-making builds on this foundation by applying the insights from reasoning to select the best course of action among alternatives. Together, these processes create a continuous cycle of thought and action aimed at achieving goals effectively. As AI technology evolves, there is a growing trend to train LLMs to excel in general reasoning. This study explores how the general reasoning capabilities of LLMs connect to their performance in domain-specific reasoning tasks.

cs.CL↗

arXiv2025

CXR-LT 2024: A MICCAI challenge on long-tailed, multi-label, and zero-shot disease classification from chest X-ray

The CXR-LT series is a community-driven initiative designed to enhance lung disease classification using chest X-rays (CXR). It tackles challenges in open long-tailed lung disease classification and enhances the measurability of state-of-the-art techniques. The first event, CXR-LT 2023, aimed to achieve these goals by providing high-quality benchmark CXR data for model development and conducting comprehensive evaluations to identify ongoing issues impacting lung disease classification performance. Building on the success of CXR-LT 2023, the CXR-LT 2024 expands the dataset to 377,110 chest X-rays (CXRs) and 45 disease labels, including 19 new rare disease findings. It also introduces a new focus on zero-shot learning to address limitations identified in the previous event. Specifically, CXR-LT 2024 features three tasks: (i) long-tailed classification on a large, noisy test set, (ii) long-tailed classification on a manually annotated "gold standard" subset, and (iii) zero-shot generalization to five previously unseen disease findings. This paper provides an overview of CXR-LT 2024, detailing the data curation process and consolidating state-of-the-art solutions, including the use of multimodal models for rare disease detection, advanced generative approaches to handle noisy labels, and zero-shot learning strategies for unseen diseases. Additionally, the expanded dataset enhances disease coverage to better represent real-world clinical settings, offering a valuable resource for future research. By synthesizing the insights and innovations of participating teams, we aim to advance the development of clinically realistic and generalizable diagnostic models for chest radiography.

cs.CV↗