arXiv subjects

Hui Wei

Publications and source records attributed to Hui Wei.

At least 37 records · Page 2Linked to original sources

arXiv2025

PlanGenLLMs: A Modern Survey of LLM Planning Capabilities

LLMs have immense potential for generating plans, transforming an initial world state into a desired goal state. A large body of research has explored the use of LLMs for various planning tasks, from web navigation to travel planning and database querying. However, many of these systems are tailored to specific problems, making it challenging to compare them or determine the best approach for new tasks. There is also a lack of clear and consistent evaluation criteria. Our survey aims to offer a comprehensive overview of current LLM planners to fill this gap. It builds on foundational work by Kartam and Wilkins (1990) and examines six key performance criteria: completeness, executability, optimality, representation, generalization, and efficiency. For each, we provide a thorough analysis of representative works and highlight their strengths and weaknesses. Our paper also identifies crucial future directions, making it a valuable resource for both practitioners and newcomers interested in leveraging LLM planning to support agentic workflows.

cs.AI↗

arXiv2025

Graph-Based Physics-Guided Urban PM2.5 Air Quality Imputation with Constrained Monitoring Data

This work introduces GraPhy, a graph-based, physics-guided learning framework for high-resolution and accurate air quality modeling in urban areas with limited monitoring data. Fine-grained air quality monitoring information is essential for reducing public exposure to pollutants. However, monitoring networks are often sparse in socioeconomically disadvantaged regions, limiting the accuracy and resolution of air quality modeling. To address this, we propose a physics-guided graph neural network architecture called GraPhy with layers and edge features designed specifically for low-resolution monitoring data. Experiments using data from California's socioeconomically disadvantaged San Joaquin Valley show that GraPhy achieves the overall best performance evaluated by mean squared error (MSE), mean absolute error (MAE), and R-square value (R2), improving the performance by 9%-56% compared to various baseline models. Moreover, GraPhy consistently outperforms baselines across different spatial heterogeneity levels, demonstrating the effectiveness of our model design.

cs.LG↗

arXiv2025

Response-Level Rewards Are All You Need for Online Reinforcement Learning in LLMs: A Mathematical Perspective

We study a common challenge in reinforcement learning for large language models (LLMs): the Zero-Reward Assumption, where non-terminal actions (i.e., intermediate token generations) receive zero task-specific immediate reward, while only the final token receives a reward for the entire response. This assumption arises frequently in practice, as precise token-level rewards are often difficult or infeasible to obtain in LLM applications. In this work, we provide a unifying theoretical perspective. We introduce the Trajectory Policy Gradient Theorem, which shows that the policy gradient based on true, unknown token-level rewards can be unbiasedly estimated using only a response-level reward model, regardless of whether the Zero-Reward Assumption holds or not, for algorithms in the REINFORCE and Actor-Critic families. This result reveals that widely used methods such as PPO, GRPO, ReMax, and RLOO inherently possess the capacity to model token-level reward signals, offering a theoretical justification for response-level reward approaches. Our findings pave the way for more practical, efficient LLM fine-tuning, allowing developers to treat training algorithms as black boxes and focus on improving the response-level reward model with auxiliary sub-models. We also offer a detailed analysis of popular RL and non-RL methods, comparing their theoretical foundations and practical advantages across common LLM tasks. Finally, we propose a new algorithm: Token-Reinforced Policy Optimization (TRePO), a theoretically grounded method that is simpler than PPO, matches GRPO in memory efficiency, and holds promise for broad applicability.

cs.LG↗

arXiv2025

Systematic Evaluation of LLM-as-a-Judge in LLM Alignment Tasks: Explainable Metrics and Diverse Prompt Templates

LLM-as-a-Judge has been widely applied to evaluate and compare different LLM alignmnet approaches (e.g., RLHF and DPO). However, concerns regarding its reliability have emerged, due to LLM judges' biases and inconsistent decision-making. Previous research has developed evaluation frameworks to assess reliability of LLM judges and their alignment with human preferences. However, the employed evaluation metrics often lack adequate explainability and fail to address LLM internal inconsistency. Additionally, existing studies inadequately explore the impact of various prompt templates when applying LLM-as-a-Judge methods, leading to potentially inconsistent comparisons between different alignment algorithms. In this work, we systematically evaluate LLM-as-a-Judge on alignment tasks by defining more theoretically interpretable evaluation metrics and explicitly mitigating LLM internal inconsistency from reliability metrics. We develop an open-source framework to evaluate, compare, and visualize the reliability and alignment of LLM judges, which facilitates practitioners to choose LLM judges for alignment tasks. In the experiments, we examine effects of diverse prompt templates on LLM-judge reliability and also demonstrate our developed framework by comparing various LLM judges on two common alignment datasets (i.e., TL;DR Summarization and HH-RLHF-Helpfulness). Our results indicate a significant impact of prompt templates on LLM judge performance, as well as a mediocre alignment level between the tested LLM judges and human evaluators.

cs.CL↗

arXiv2024

Local Features Meet Stochastic Anonymization: Revolutionizing Privacy-Preserving Face Recognition for Black-Box Models

The task of privacy-preserving face recognition (PPFR) currently faces two major unsolved challenges: (1) existing methods are typically effective only on specific face recognition models and struggle to generalize to black-box face recognition models; (2) current methods employ data-driven reversible representation encoding for privacy protection, making them susceptible to adversarial learning and reconstruction of the original image. We observe that face recognition models primarily rely on local features ({e.g., face contour, skin texture, and so on) for identification. Thus, by disrupting global features while enhancing local features, we achieve effective recognition even in black-box environments. Additionally, to prevent adversarial models from learning and reversing the anonymization process, we adopt an adversarial learning-based approach with irreversible stochastic injection to ensure the stochastic nature of the anonymization. Experimental results demonstrate that our method achieves an average recognition accuracy of 94.21\% on black-box models, outperforming existing methods in both privacy protection and anti-reconstruction capabilities.

cs.CV↗

arXiv2024

Physical Adversarial Attack meets Computer Vision: A Decade Survey

Despite the impressive achievements of Deep Neural Networks (DNNs) in computer vision, their vulnerability to adversarial attacks remains a critical concern. Extensive research has demonstrated that incorporating sophisticated perturbations into input images can lead to a catastrophic degradation in DNNs' performance. This perplexing phenomenon not only exists in the digital space but also in the physical world. Consequently, it becomes imperative to evaluate the security of DNNs-based systems to ensure their safe deployment in real-world scenarios, particularly in security-sensitive applications. To facilitate a profound understanding of this topic, this paper presents a comprehensive overview of physical adversarial attacks. Firstly, we distill four general steps for launching physical adversarial attacks. Building upon this foundation, we uncover the pervasive role of artifacts carrying adversarial perturbations in the physical world. These artifacts influence each step. To denote them, we introduce a new term: adversarial medium. Then, we take the first step to systematically evaluate the performance of physical adversarial attacks, taking the adversarial medium as a first attempt. Our proposed evaluation metric, hiPAA, comprises six perspectives: Effectiveness, Stealthiness, Robustness, Practicability, Aesthetics, and Economics. We also provide comparative results across task categories, together with insightful observations and suggestions for future research directions.

cs.CV↗

arXiv2024

Why Does the Cortex Have Such a Vast Storage Capacity?

The capacity of long-term memory seems to be extremely large, capable of storing information spanning almost a lifetime. Why does it have such a vast capacity? Why are some memories so enduring? What is the actual physical form of long-term memory? In the movie Inside Out, it is depicted as individual orbs containing information. Is that really the case? Simply explaining this by saying that the cortex has many neurons, numerous neural connections, and complex electrochemical activity between them is not sufficient to answer these fundamental questions. We need to uncover the theory hidden behind these phenomena.In essence, a neural network is equivalent to a very large directed graph, with a massive number of nodes and directed connections. This paper posits that the physical form of long-term memory is a connected subgraph within this complex directed graph. This subgraph is capable of linking together the disparate fragments of the same event, spread across different sensory cortices, to form associations. This provides a physical realization of the engram theory. The robustness of the connected subgraph and the resources it consumes can explain various memory behaviors.Based on anatomical, brain imaging and electrophysiological evidence, this paper constructs a probabilistic connectivity model and uses theorems from graph theory to prove the ease of constructing connected subgraphs. Finally, it explains why the potential capacity for memory is immense.

q-bio.NC↗

arXiv2024

Integrating One-Shot View Planning with a Single Next-Best View via Long-Tail Multiview Sampling

Existing view planning systems either adopt an iterative paradigm using next-best views (NBV) or a one-shot pipeline relying on the set-covering view-planning (SCVP) network. However, neither of these methods can concurrently guarantee both high-quality and high-efficiency reconstruction of 3D unknown objects. To tackle this challenge, we introduce a crucial hypothesis: with the availability of more information about the unknown object, the prediction quality of the SCVP network improves. There are two ways to provide extra information: (1) leveraging perception data obtained from NBVs, and (2) training on an expanded dataset of multiview inputs. In this work, we introduce a novel combined pipeline that incorporates a single NBV before activating the proposed multiview-activated (MA-)SCVP network. The MA-SCVP is trained on a multiview dataset generated by our long-tail sampling method, which addresses the issue of unbalanced multiview inputs and enhances the network performance. Extensive simulated experiments substantiate that our system demonstrates a significant surface coverage increase and a substantial 45% reduction in movement cost compared to state-of-the-art systems. Real-world experiments justify the capability of our system for generalization and deployment.

cs.RO↗

arXiv2024

REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learning

The success of self-supervised contrastive learning hinges on identifying positive data pairs, such that when they are pushed together in embedding space, the space encodes useful information for subsequent downstream tasks. Constructing positive pairs is non-trivial as the pairing must be similar enough to reflect a shared semantic meaning, but different enough to capture within-class variation. Classical approaches in vision use augmentations to exploit well-established invariances to construct positive pairs, but invariances in the time-series domain are much less obvious. In our work, we propose a novel method of using a learned measure for identifying positive pairs. Our Retrieval-Based Reconstruction (REBAR) measure measures the similarity between two sequences as the reconstruction error that results from reconstructing one sequence with retrieved information from the other. Then, if the two sequences have high REBAR similarity, we label them as a positive pair. Through validation experiments, we show that the REBAR error is a predictor of mutual class membership. Once integrated into a contrastive learning framework, our REBAR method learns an embedding that achieves state-of-the-art performance on downstream tasks across various modalities.

cs.LG↗

arXiv2024

Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation

Wearable sensors enable health researchers to continuously collect data pertaining to the physiological state of individuals in real-world settings. However, such data can be subject to extensive missingness due to a complex combination of factors. In this work, we study the problem of imputation of missing step count data, one of the most ubiquitous forms of wearable sensor data. We construct a novel and large scale data set consisting of a training set with over 3 million hourly step count observations and a test set with over 2.5 million hourly step count observations. We propose a domain knowledge-informed sparse self-attention model for this task that captures the temporal multi-scale nature of step-count data. We assess the performance of the model relative to baselines and conduct ablation studies to verify our specific model designs.

cs.LG↗

arXiv2024

Modeling of Memory Mechanisms in Cerebral Cortex and Simulation of Storage Performance

At the intersection of computation and cognitive science, graph theory is utilized as a formalized description of complex relationships and structures. Traditional graph models are often static, lacking dynamic and autonomous behavioral patterns. They rely on algorithms with a global view, significantly differing from biological neural networks, in which, to simulate information storage and retrieval processes, the limitations of centralized algorithms must be overcome. This study introduces a directed graph model that equips each node with adaptive learning and decision-making capabilities, thereby facilitating decentralized dynamic information storage and modeling and simulation of the brain's memory process. We abstract different storage instances as directed graph paths, transforming the storage of information into the assignment, discrimination, and extraction of different paths. To address writing and reading challenges, each node has a personalized adaptive learning ability. A storage algorithm without a God's eye view is developed, where each node uses its limited neighborhood information to facilitate the extension, formation, solidification, and awakening of directed graph paths, achieving competitive, reciprocal, and sustainable utilization of limited resources. Storage behavior occurs in each node, with adaptive learning behaviors of nodes concretized in a microcircuit centered around a variable resistor, simulating the electrophysiological behavior of neurons. Under the constraints of neurobiology on the anatomy and electrophysiology of biological neural networks, this model offers a plausible explanation for the mechanism of memory realization, providing a comprehensive, system-level experimental validation of the memory trace theory.

q-bio.NC↗

arXiv2024

On Support Relations Inference and Scene Hierarchy Graph Construction from Point Cloud in Clustered Environments

Over the years, scene understanding has attracted a growing interest in computer vision, providing the semantic and physical scene information necessary for robots to complete some particular tasks autonomously. In 3D scenes, rich spatial geometric and topological information are often ignored by RGB-based approaches for scene understanding. In this study, we develop a bottom-up approach for scene understanding that infers support relations between objects from a point cloud. Our approach utilizes the spatial topology information of the plane pairs in the scene, consisting of three major steps. 1) Detection of pairwise spatial configuration: dividing primitive pairs into local support connection and local inner connection; 2) primitive classification: a combinatorial optimization method applied to classify primitives; and 3) support relations inference and hierarchy graph construction: bottom-up support relations inference and scene hierarchy graph construction containing primitive level and object level. Through experiments, we demonstrate that the algorithm achieves excellent performance in primitive classification and support relations inference. Additionally, we show that the scene hierarchy graph contains rich geometric and topological information of objects, and it possesses great scalability for scene understanding.

cs.CV↗

arXiv2023

Autonomous and Ubiquitous In-node Learning Algorithms of Active Directed Graphs and Its Storage Behavior

Memory is an important cognitive function for humans. How a brain with such a small power can complete such a complex memory function, the working mechanism behind this is undoubtedly fascinating. Engram theory views memory as the co-activation of specific neuronal clusters. From the perspective of graph theory, nodes represent neurons, and directed edges represent synapses. Then the memory engram is the connected subgraph formed between the activated nodes. In this paper, we use subgraphs as physical carriers of information and propose a parallel distributed information storage algorithm based on node scale in active-directed graphs. An active-directed graph is defined as a graph in which each node has autonomous and independent behavior and relies only on information obtained within the local field of view to make decisions. Unlike static directed graphs used for recording facts, active-directed graphs are decentralized like biological neuron networks and do not have a super manager who has a global view and can control the behavior of each node. Distinct from traditional algorithms with a global field of view, this algorithm is characterized by nodes collaborating globally on resource usage through their limited local field of view. While this strategy may not achieve global optimality as well as algorithms with a global field of view, it offers better robustness, concurrency, decentralization, and bioviability. Finally, it was tested in network capacity, fault tolerance, and robustness. It was found that the algorithm exhibits a larger network capacity in a more sparse network structure because the subgraph generated by a single sample is not a whole but consists of multiple weakly connected components. In this case, the network capacity can be understood as the number of permutations of several weakly connected components in the network.

cs.DC↗

arXiv2023

Important Clues that Facilitate Visual Emergence: Three Psychological Experiments

Visual emergence is the phenomenon in which the visual system obtains a holistic perception after grouping and reorganizing local signals. The picture Dalmatian dog is known for its use in explaining visual emergence. This type of image, which consists of a set of discrete black speckles (speckles), is called an emerging image. Not everyone can find the dog in Dalmatian dog, and among those who can, the time spent varies greatly. Although Gestalt theory summarizes perceptual organization into several principles, it remains ambiguous how these principles affect the perception of emerging images. This study, therefore, designed three psychological experiments to explore the factors that influence the perception of emerging images. In the first, we found that the density of speckles in the local area and the arrangements of some key speckles played a key role in the perception of an emerging case. We set parameters in the algorithm to characterize these two factors. We then automatically generated diversified emerging-test images (ETIs) through the algorithm and verified their effectiveness in two subsequent experiments.

q-bio.NC↗

arXiv2023

Object Detection based on the Collection of Geometric Evidence

Artificial objects usually have very stable shape features, which are stable, persistent properties in geometry. They can provide evidence for object recognition. Shape features are more stable and more distinguishing than appearance features, color features, grayscale features, or gradient features. The difficulty with object recognition based on shape features is that objects may differ in color, lighting, size, position, pose, and background interference, and it is not currently possible to predict all possible conditions. The variety of objects and conditions renders object recognition based on geometric features very challenging. This paper provides a method based on shape templates, which involves the selection, collection, and combination discrimination of geometric evidence of the edge segments of images, to find out the target object accurately from background, and it is able to identify the semantic attributes of each line segment of the target object. In essence, the method involves solving a global optimal combinatorial optimization problem. Although the complexity of the global optimal combinatorial optimization problem seems to be very high, there is no need to define the complex feature vector and no need for any expensive training process. It has very good generalization ability and environmental adaptability, and more solid basis for cognitive psychology than other methods. The process of collecting geometric evidence, which is simple and universal, shows considerable prospects for practical use. The experimental results prove that the method has great advantages in response to changes in the environment, invariant recognition, pinpointing the geometry of objects, search efficiency, and efficient calculation. This attempt contributes to understanding of some types of universal processing during the process of object recognition.

cs.CV↗

arXiv2022

HOTCOLD Block: Fooling Thermal Infrared Detectors with a Novel Wearable Design

Adversarial attacks on thermal infrared imaging expose the risk of related applications. Estimating the security of these systems is essential for safely deploying them in the real world. In many cases, realizing the attacks in the physical space requires elaborate special perturbations. These solutions are often \emph{impractical} and \emph{attention-grabbing}. To address the need for a physically practical and stealthy adversarial attack, we introduce \textsc{HotCold} Block, a novel physical attack for infrared detectors that hide persons utilizing the wearable Warming Paste and Cooling Paste. By attaching these readily available temperature-controlled materials to the body, \textsc{HotCold} Block evades human eyes efficiently. Moreover, unlike existing methods that build adversarial patches with complex texture and structure features, \textsc{HotCold} Block utilizes an SSP-oriented adversarial optimization algorithm that enables attacks with pure color blocks and explores the influence of size, shape, and position on attack performance. Extensive experimental results in both digital and physical environments demonstrate the performance of our proposed \textsc{HotCold} Block. \emph{Code is available: \textcolor{magenta}{https://github.com/weihui1308/HOTCOLDBlock}}.

cs.CV↗

arXiv2021

Periodic solutions of a semilinear variable coefficient wave equation under asymptotic nonresonance conditions

We consider the periodic solutions of a semilinear variable coefficient wave equation arising from the forced vibrations of a nonhomogeneous string and the propagation of seismic waves in nonisotropic media. The variable coefficient characterizes the inhomogeneity of media and its presence usually leads to the destruction of the compactness of the inverse of linear wave operator with periodic-Dirichlet boundary conditions on its range. In the pioneering work of Barbu and Pavel (1997), it gives the existence and regularity of periodic solution for Lipschitz, nonresonant and monotone nonlinearity under the assumption $η_u>0$ (see Sect. 2 for its definition) on the coefficient $u(x)$ and leaves the case $η_u=0$ as an open problem. In this paper, by developing the invariant subspace method and using the complete reduction technique and Leray-Schauder theory, we obtain the existence of periodic solutions for such a problem when the nonlinear term satisfies the asymptotic nonresonance conditions. Our result not only does not need any requirements on the coefficient except for the natural positivity assumption (i.e., $u(x)>0$), but also does not need the monotonicity assumption on the nonlinearity. In particular, when the nonlinear term is an odd function and satisfies the global nonresonance conditions, there is only one (trivial) solution to this problem on the invariant subspace.

math.AP↗

arXiv2021

Probability Trajectory: One New Movement Description for Trajectory Prediction

Trajectory prediction is a fundamental and challenging task for numerous applications, such as autonomous driving and intelligent robots. Currently, most of existing work treat the pedestrian trajectory as a series of fixed two-dimensional coordinates. However, in real scenarios, the trajectory often exhibits randomness, and has its own probability distribution. Inspired by this observed fact, also considering other movement characteristics of pedestrians, we propose one simple and intuitive movement description, probability trajectory, which maps the coordinate points of pedestrian trajectory into two-dimensional Gaussian distribution in images. Based on this unique description, we develop one novel trajectory prediction method, called social probability. The method combines the new probability trajectory and powerful convolution recurrent neural networks together. Both the input and output of our method are probability trajectories, which provide the recurrent neural network with sufficient spatial and random information of moving pedestrians. And the social probability extracts spatio-temporal features directly on the new movement description to generate robust and accurate predicted results. The experiments on public benchmark datasets show the effectiveness of the proposed method.

cs.CV↗