arXiv subjects

Xiaoqiang Wang

Publications and source records attributed to Xiaoqiang Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems

The advent of large language models (LLMs) has catalyzed a transformative shift in artificial intelligence, paving the way for advanced intelligent agents capable of sophisticated reasoning, robust perception, and versatile action across diverse domains. As these agents increasingly drive AI research and practical applications, their design, evaluation, and continuous improvement present intricate, multifaceted challenges. This book provides a comprehensive overview, framing intelligent agents within modular, brain-inspired architectures that integrate principles from cognitive science, neuroscience, and computational research. We structure our exploration into four interconnected parts. First, we systematically investigate the modular foundation of intelligent agents, systematically mapping their cognitive, perceptual, and operational modules onto analogous human brain functionalities and elucidating core components such as memory, world modeling, reward processing, goal, and emotion. Second, we discuss self-enhancement and adaptive evolution mechanisms, exploring how agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual learning through automated optimization paradigms. Third, we examine multi-agent systems, investigating the collective intelligence emerging from agent interactions, cooperation, and societal structures. Finally, we address the critical imperative of building safe and beneficial AI systems, emphasizing intrinsic and extrinsic security threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy real-world deployment. By synthesizing modular AI architectures with insights from different disciplines, this survey identifies key research challenges and opportunities, encouraging innovations that harmonize technological advancement with meaningful societal benefit.

cs.AI↗

arXiv2025

Accelerating PDE-Constrained Optimization by the Derivative of Neural Operators

PDE-Constrained Optimization (PDECO) problems can be accelerated significantly by employing gradient-based methods with surrogate models like neural operators compared to traditional numerical solvers. However, this approach faces two key challenges: (1) **Data inefficiency**: Lack of efficient data sampling and effective training for neural operators, particularly for optimization purpose. (2) **Instability**: High risk of optimization derailment due to inaccurate neural operator predictions and gradients. To address these challenges, we propose a novel framework: (1) **Optimization-oriented training**: we leverage data from full steps of traditional optimization algorithms and employ a specialized training method for neural operators. (2) **Enhanced derivative learning**: We introduce a *Virtual-Fourier* layer to enhance derivative learning within the neural operator, a crucial aspect for gradient-based optimization. (3) **Hybrid optimization**: We implement a hybrid approach that integrates neural operators with numerical solvers, providing robust regularization for the optimization process. Our extensive experimental results demonstrate the effectiveness of our model in accurately learning operators and their derivatives. Furthermore, our hybrid optimization approach exhibits robust convergence.

cs.LG↗

arXiv2025

System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts

Chain-of-thought (CoT) reasoning enables large language models (LLMs) to move beyond fast System-1 responses and engage in deliberative System-2 reasoning. However, this comes at the cost of significant inefficiency due to verbose intermediate output. Recent latent-space reasoning methods improve efficiency by operating on hidden states without decoding into language, yet they treat all steps uniformly, failing to distinguish critical deductions from auxiliary steps and resulting in suboptimal use of computational resources. In this paper, we propose System-1.5 Reasoning, an adaptive reasoning framework that dynamically allocates computation across reasoning steps through shortcut paths in latent space. Specifically, System-1.5 Reasoning introduces two types of dynamic shortcuts. The model depth shortcut (DS) adaptively reasons along the vertical depth by early exiting non-critical tokens through lightweight adapter branches, while allowing critical tokens to continue through deeper Transformer layers. The step shortcut (SS) reuses hidden states across the decoding steps to skip trivial steps and reason horizontally in latent space. Training System-1.5 Reasoning involves a two-stage self-distillation process: first distilling natural language CoT into latent-space continuous thought, and then distilling full-path System-2 latent reasoning into adaptive shortcut paths (System-1.5 Reasoning). Experiments on reasoning tasks demonstrate the superior performance of our method. For example, on GSM8K, System-1.5 Reasoning achieves reasoning performance comparable to traditional CoT fine-tuning methods while accelerating inference by over 20x and reducing token generation by 92.31% on average.

cs.CL↗

arXiv2025

Improvement of the square-root low bounds on the minimum distances of BCH codes and Matrix-product codes

The task of constructing infinite families of self-dual codes with unbounded lengths and minimum distances exhibiting square-root lower bounds is extremely challenging, especially when it comes to cyclic codes. Recently, the first infinite family of Euclidean self-dual binary and nonbinary cyclic codes, whose minimum distances have a square-root lower bound and have a lower bound better than square-root lower bounds are constructed in \cite{Chen23} for the lengths of these codes being unbounded. Let $q$ be a power of a prime number and $Q=q^2$. In this paper, we first improve the lower bounds on the minimum distances of Euclidean and Hermitian duals of BCH codes with length $\frac{q^m-1}{q^s-1}$ over $\mathbb{F}_q$ and $\frac{Q^m-1}{Q-1}$ over $\mathbb{F}_Q$ in \cite{Fan23,GDL21,Wang24} for the designed distances in some ranges, respectively, where $\frac{m}{s}\geq 3$. Then based on matrix-product construction and some lower bounds on the minimum distances of BCH codes and their duals, we obtain several classes of Euclidean and Hermitian self-dual codes, whose minimum distances have square-root lower bounds or a square-root-like lower bounds. Our lower bounds on the minimum distances of Euclidean and Hermitian self-dual cyclic codes improved many results in \cite{Chen23}. In addition, our lower bounds on the minimum distances of the duals of BCH codes are almost $q^s-1$ or $q$ times that of the existing lower bounds.

cs.IT↗

arXiv2025

Several new infinite families of NMDS codes with arbitrary dimensions supporting $t$-designs

Near maximum distance separable (NMDS) codes, where both the code and its dual are almost maximum distance separable, play pivotal roles in combinatorial design theory and cryptographic applications. Despite progress in fixed dimensions (e.g., dimension 4 codes by Ding and Tang \cite{Ding2020}), constructing NMDS codes with arbitrary dimensions supporting $t$-designs ($t\geq 2$) has remained open. In this paper, we construct two infinite families of NMDS codes over $\mathbb{F}_q$ for any prime power $q$ with flexible dimensions and determine their weight distributions. Further, two additional families with arbitrary dimensions over $\mathbb{F}_{2^m}$ supporting $2$-designs and $3$-designs, and their weight distributions are obtained. Our results fully generalize prior fixed-dimension works~\cite{DingY2024,Heng2023,Heng20231,Xu2022}, and affirmatively settle the Heng-Wang conjecture \cite{Heng2023} on the existence of NMDS codes with flexible parameters supporting $2$-designs.

cs.IT↗

arXiv2025

R$^3$Mem: Bridging Memory Retention and Retrieval via Reversible Compression

Memory plays a key role in enhancing LLMs' performance when deployed to real-world applications. Existing solutions face trade-offs: explicit memory designs based on external storage require complex management and incur storage overhead, while implicit memory designs that store information via parameters struggle with reliable retrieval. In this paper, we propose R$^3$Mem, a memory network that optimizes both information Retention and Retrieval through Reversible context compression. Specifically, R$^3$Mem employs virtual memory tokens to compress and encode infinitely long histories, further enhanced by a hierarchical compression strategy that refines information from document- to entity-level for improved assimilation across granularities. For retrieval, R$^3$Mem employs a reversible architecture, reconstructing raw data by invoking the model backward with compressed information. Implemented via parameter-efficient fine-tuning, it can integrate seamlessly with any Transformer-based model. Experiments demonstrate that our memory design achieves state-of-the-art performance in long-context language modeling and retrieval-augmented generation tasks. It also significantly outperforms conventional memory modules in long-horizon interaction tasks like conversational agents, showcasing its potential for next-generation retrieval systems.

cs.CL↗

arXiv2024

A new class of self-orthogonal linear codes and their applications

Self-orthogonal codes are a subclass of linear codes that are contained within their dual codes. Since self-orthogonal codes are widely used in quantum codes, lattice theory and linear complementary dual (LCD) codes, they have received continuous attention and research. In this paper, we construct a class of self-orthogonal codes by using the defining-set approach, and determine their explicit weight distributions and the parameters of their duals. Some considered codes are optimal according to the tables of best codes known maintained at \cite{Grassl} and a class of almost maximum distance separable (AMDS) codes from their duals are obtained. As applications, we obtain a class of new quantum codes, which are MDS or AMDS according to the quantum Singleton bound under certain conditions. Some examples show that the constructed quantum codes have the better parameters than known ones maintained at \cite{Bierbrauer}. Furthermore, a new class of LCD codes are given, which are almost optimal according to the sphere packing bound.

cs.IT↗

arXiv2024

OSCAR: Operating System Control via State-Aware Reasoning and Re-Planning

Large language models (LLMs) and large multimodal models (LMMs) have shown great potential in automating complex tasks like web browsing and gaming. However, their ability to generalize across diverse applications remains limited, hindering broader utility. To address this challenge, we present OSCAR: Operating System Control via state-Aware reasoning and Re-planning. OSCAR is a generalist agent designed to autonomously navigate and interact with various desktop and mobile applications through standardized controls, such as mouse and keyboard inputs, while processing screen images to fulfill user commands. OSCAR translates human instructions into executable Python code, enabling precise control over graphical user interfaces (GUIs). To enhance stability and adaptability, OSCAR operates as a state machine, equipped with error-handling mechanisms and dynamic task re-planning, allowing it to efficiently adjust to real-time feedback and exceptions. We demonstrate OSCAR's effectiveness through extensive experiments on diverse benchmarks across desktop and mobile platforms, where it transforms complex workflows into simple natural language commands, significantly boosting user productivity. Our code will be open-source upon publication.

cs.AI↗

arXiv2024

FAC$^2$E: Better Understanding Large Language Model Capabilities by Dissociating Language and Cognition

Large language models (LLMs) are primarily evaluated by overall performance on various text understanding and generation tasks. However, such a paradigm fails to comprehensively differentiate the fine-grained language and cognitive skills, rendering the lack of sufficient interpretation to LLMs' capabilities. In this paper, we present FAC$^2$E, a framework for Fine-grAined and Cognition-grounded LLMs' Capability Evaluation. Specifically, we formulate LLMs' evaluation in a multi-dimensional and explainable manner by dissociating the language-related capabilities and the cognition-related ones. Besides, through extracting the intermediate reasoning from LLMs, we further break down the process of applying a specific capability into three sub-steps: recalling relevant knowledge, utilizing knowledge, and solving problems. Finally, FAC$^2$E evaluates each sub-step of each fine-grained capability, providing a two-faceted diagnosis for LLMs. Utilizing FAC$^2$E, we identify a common shortfall in knowledge utilization among models and propose a straightforward, knowledge-enhanced method to mitigate this issue. Our results not only showcase promising performance enhancements but also highlight a direction for future LLM advancements.

cs.CL↗

arXiv2024

The weight hierarchies of three classes of linear codes

Studying the generalized Hamming weights of linear codes is a significant research area within coding theory, as it provides valuable structural information about the codes and plays a crucial role in determining their performance in various applications. However, determining the generalized Hamming weights of linear codes, particularly their weight hierarchy, is generally a challenging task. In this paper, we focus on investigating the generalized Hamming weights of three classes of linear codes over finite fields. These codes are constructed by different defining sets. By analysing the intersections between the definition sets and the duals of all $r$-dimensional subspaces, we get the inequalities on the sizes of these intersections. Then constructing subspaces that reach the upper bounds of these inequalities, we successfully determine the complete weight hierarchies of these codes.

cs.IT↗

arXiv2024

A Mallows-like Criterion for Anomaly Detection with Random Forest Implementation

The effectiveness of anomaly signal detection can be significantly undermined by the inherent uncertainty of relying on one specified model. Under the framework of model average methods, this paper proposes a novel criterion to select the weights on aggregation of multiple models, wherein the focal loss function accounts for the classification of extremely imbalanced data. This strategy is further integrated into Random Forest algorithm by replacing the conventional voting method. We have evaluated the proposed method on benchmark datasets across various domains, including network intrusion. The findings indicate that our proposed method not only surpasses the model averaging with typical loss functions but also outstrips common anomaly detection algorithms in terms of accuracy and robustness.

stat.ML↗

arXiv2024

Reference Neural Operators: Learning the Smooth Dependence of Solutions of PDEs on Geometric Deformations

For partial differential equations on domains of arbitrary shapes, existing works of neural operators attempt to learn a mapping from geometries to solutions. It often requires a large dataset of geometry-solution pairs in order to obtain a sufficiently accurate neural operator. However, for many industrial applications, e.g., engineering design optimization, it can be prohibitive to satisfy the requirement since even a single simulation may take hours or days of computation. To address this issue, we propose reference neural operators (RNO), a novel way of implementing neural operators, i.e., to learn the smooth dependence of solutions on geometric deformations. Specifically, given a reference solution, RNO can predict solutions corresponding to arbitrary deformations of the referred geometry. This approach turns out to be much more data efficient. Through extensive experiments, we show that RNO can learn the dependence across various types and different numbers of geometry objects with relatively small datasets. RNO outperforms baseline models in accuracy by a large lead and achieves up to 80% error reduction.

cs.LG↗

arXiv2023

The duals of narrow-sense BCH codes with length $\frac{q^m-1}λ$

BCH codes are an interesting class of cyclic codes due to their efficient encoding and decoding algorithms. In the past sixty years, a lot of progress on the study of BCH codes has been made, but little is known about the properties of their duals. Recently, in order to study the duals of BCH codes and the lower bounds on their minimum distances, a new concept called dually-BCH code was proposed by authors in \cite{GDL21}. In this paper, the lower bounds on the minimum distances of the duals of narrow-sense BCH codes with length $\frac{q^m-1}λ$ over $\mathbb{F}_q$ are developed, where $λ$ is a positive integer satisfying $λ\, |\, q-1$, or $λ=q^s-1$ and $s\, |\,m$. In addition, the sufficient and necessary conditions in terms of the designed distances for these codes being dually-BCH codes are presented. Many considered codes in \cite{GDL21} and \cite{Wang23} are the special cases of the codes showed in this paper. Our lower bounds on the minimum distances of the duals of BCH codes include the bounds stated in \cite{GDL21} as a special case. Several examples show that the lower bounds are good in some cases.

cs.IT↗

arXiv2023

The Weight Hierarchies of Linear Codes from Simplicial Complexes

The study of the generalized Hamming weight of linear codes is a significant research topic in coding theory as it conveys the structural information of the codes and determines their performance in various applications. However, determining the generalized Hamming weights of linear codes, especially the weight hierarchy, is generally challenging. In this paper, we investigate the generalized Hamming weights of a class of linear code $\C$ over $\bF_q$, which is constructed from defining sets. These defining sets are either special simplicial complexes or their complements in $\bF_q^m$. We determine the complete weight hierarchies of these codes by analyzing the maximum or minimum intersection of certain simplicial complexes and all $r$-dimensional subspaces of $\bF_q^m$, where $1\leq r\leq {\rm dim}_{\bF_q}(\C)$.

cs.IT↗

arXiv2023

k-dependent proximity-induced modulation of spin-orbit interaction in MoSe2 interfaced with amorphous Pb

The ability to modulate the spin-orbit (SO) interaction is crucial for engineering a wide range of spintronics-based quantum devices, extending from state-of-the-art data storage to materials for quantum computing. The use of proximity-induced effects for this purpose may become the mainstream approach, whereas their experimental verification using angle-resolved photoelectron spectroscopy (ARPES) has so far been elusive. Here, using the advantages of soft-X-ray ARPES on its probing depth and intrinsic resolution in three-dimensional momentum k, we identify a distinct modulation of the SO interaction in a van der Waals semiconductor (MoSe2) proximitized to a high-Z metal (Pb), and measure its variation through the k-space. The strong SO field from Pb boosts the SO splitting by up to 30% at the H-point of the bulk Brillouin zone, the spin-orbit hotspot of MoSe2. Tunability of the splitting via the Pb thickness allows its tailoring to particular applications in emerging quantum devices.

cond-mat.mtrl-sci↗

arXiv2023

Two classes of reducible cyclic codes with large minimum symbol-pair distances

The high-density data storage technology aims to design high-capacity storage at a relatively low cost. In order to achieve this goal, symbol-pair codes were proposed by Cassuto and Blaum \cite{CB10,CB11} to handle channels that output pairs of overlapping symbols. Such a channel is called symbol-pair read channel, which introduce new concept called symbol-pair weight and minimum symbol-pair distance. In this paper, we consider the parameters of two classes of reducible cyclic codes under the symbol-pair metric. Based on the theory of cyclotomic numbers and Gaussian period over finite fields, we show the possible symbol-pair weights of these codes. Their minimum symbol-pair distances are twice the minimum Hamming distances under some conditions. Moreover, we obtain some three symbol-pair weight codes and determine their symbol-pair weight distribution. A class of MDS symbol-pair codes is also established. Among other results, we determine the values of some generalized cyclotomic numbers.

cs.IT↗

arXiv2023

SkillQG: Learning to Generate Question for Reading Comprehension Assessment

We present $\textbf{$\texttt{SkillQG}$}$: a question generation framework with controllable comprehension types for assessing and improving machine reading comprehension models. Existing question generation systems widely differentiate questions by $\textit{literal}$ information such as question words and answer types to generate semantically relevant questions for a given context. However, they rarely consider the $\textit{comprehension}$ nature of questions, i.e. the different comprehension capabilities embodied by different questions. In comparison, our $\texttt{SkillQG}$ is able to tailor a fine-grained assessment and improvement to the capabilities of question answering models built on it. Specifically, we first frame the comprehension type of questions based on a hierarchical skill-based schema, then formulate $\texttt{SkillQG}$ as a skill-conditioned question generator. Furthermore, to improve the controllability of generation, we augment the input text with question focus and skill-specific knowledge, which are constructed by iteratively prompting the pre-trained language models. Empirical results demonstrate that $\texttt{SkillQG}$ outperforms baselines in terms of quality, relevance, and skill-controllability while showing a promising performance boost in downstream question answering task.

cs.CL↗

arXiv2023

Improving Contextual Spelling Correction by External Acoustics Attention and Semantic Aware Data Augmentation

We previously proposed contextual spelling correction (CSC) to correct the output of end-to-end (E2E) automatic speech recognition (ASR) models with contextual information such as name, place, etc. Although CSC has achieved reasonable improvement in the biasing problem, there are still two drawbacks for further accuracy improvement. First, due to information limitation in text only hypothesis or weak performance of ASR model on rare domains, the CSC model may fail to correct phrases with similar pronunciation or anti-context cases where all biasing phrases are not present in the utterance. Second, there is a discrepancy between the training and inference of CSC. The bias list in training is randomly selected but in inference there may be more similarity between ground truth phrase and other phrases. To solve above limitations, in this paper we propose an improved non-autoregressive (NAR) spelling correction model for contextual biasing in E2E neural transducer-based ASR systems to improve the previous CSC model from two perspectives: Firstly, we incorporate acoustics information with an external attention as well as text hypotheses into CSC to better distinguish target phrase from dissimilar or irrelevant phrases. Secondly, we design a semantic aware data augmentation schema in training phrase to reduce the mismatch between training and inference to further boost the biasing accuracy. Experiments show that the improved method outperforms the baseline ASR+Biasing system by as much as 20.3% relative name recall gain and achieves stable improvement compared to the previous CSC method over different bias list name coverage ratio.

cs.SD↗