arXiv subjects

Hongwei Liu

Publications and source records attributed to Hongwei Liu.

At least 19 recordsLinked to original sources

arXiv2026

Self-powered InAs nanowire detector arrays for extended-SWIR spectrometry at room temperature

Spectral sensing in the extended shortwave infrared (e-SWIR) is important for molecular analysis, infrared imaging, and machine vision, motivating the development of compact spectrometers for broader applications. However, conventional commercial off-the-shelf spectrometers in this wavelength region are expensive and bulky due to their reliance on external dispersive optics/filters and/or cryogenic accessories. Other emerging computational spectrometers are based on Si and InGaAs photodetectors that remain focused on the visible and near-infrared, with few detector platforms operating in the e-SWIR regime that simultaneously provide broadband sensitivity, low-noise room-temperature operation, and diverse spectral signatures for accurate identification and reconstruction. Here, we report a room-temperature e-SWIR computational spectrometer based on InAs/InP core-shell nanowire photodetector arrays with geometry-encoded spectral responses. The detectors exhibit self-powered broadband photoresponse across the 1--3 $\mu$m range, with responsivity up to 0.215 A W$^{-1}$, detectivity up to $1.6 \times 10^{9}$ cm Hz$^{1/2}$ W$^{-1}$, and microsecond response times. The excellent detector performance is leveraged to demonstrate filter-free spectral reconstruction using a compact multipixel photodetector array device. This enables high-accuracy molecular absorption spectrum reconstruction and hyperspectral imaging. Our results indicate that InAs nanowire arrays are a promising platform for compact computational spectrometry and imaging in the e-SWIR at room temperature.

physics.optics↗

arXiv2026

Function-Correcting RT Codes: A Coding Framework for Parallel Channels

Function-correcting codes have recently emerged as an effective approach to reducing coding redundancy when the objective is to reliably recover only the prescribed function values of the transmitted information, rather than the complete message itself. As a natural generalization of the Hamming metric, the Rosenbloom--Tsfasman metric (RT-metric) has attracted considerable attention due to its relevance to communication over parallel channels. Motivated by these two research directions, we introduce in this paper function-correcting RT codes (FCRTCs), which extend the framework of function-correcting codes from the Hamming metric to the more general RT metric. To investigate the fundamental problem of optimal redundancy, we introduce a new class of matrix codes, called irregular RT-distance codes, and establish a direct correspondence between the redundancy optimization problem for FCRTCs and the construction of such codes. This connection provides a unified framework for deriving redundancy bounds under the RT metric. For arbitrary functions, we establish general upper and lower bounds on the optimal redundancy of FCRTCs. We then specialize our analysis to several important classes of functions, namely RT-weight functions, RT-weight distribution functions, and RT-locally-two-valued binary functions, for which substantially sharper bounds are obtained by exploiting their specific structural properties. Furthermore, we present explicit constructions of FCRTCs for RT-weight distribution functions and prove that these constructions attain the optimal redundancy bounds in several cases. The proposed framework considerably extends the theory of function-correcting codes to the RT metric, provides new theoretical tools for studying reliable communication over parallel channels, and opens new perspectives for redundancy-efficient coding beyond the classical Hamming setting.

cs.IT↗

arXiv2026

Dual Prototype-Conditioned Diffusion Model for Scalable Multi-Class Unsupervised Anomaly Detection in Large Category Spaces

Multi-class anomaly detection aims to build unified models across diverse product categories. However, as the number of categories grows, its performance often degrades due to increasingly complex and heterogeneous normal distributions. To address this challenge, we propose DPDiff-AD, a Dual Prototype-conditioned Diffusion model for large-scale multi-class Anomaly Detection. DPDiff-AD models heterogeneous normal distributions through complementary local and global prototypes. Local prototypes capture representative fine-grained structural patterns via nearest-prototype aggregation, while global prototypes regulate holistic feature geometry through optimal transport regularization. Together, these dual-scale representations define a structured normality space. This space is refined through diffusion-based reconstruction conditioned on both local and global prototypes via prototype-aware attention. By jointly leveraging dual prototypes during generation, DPDiff-AD achieves precise normality modeling, preserves structured separability as category cardinality grows, and enables scalable anomaly discrimination. Extensive experiments across five benchmarks demonstrate the effectiveness and scalability of DPDiff-AD. On the 160-category large-scale dataset, it improves image- and pixel-level AUROC by 5.3 and 2.9 points over the previous state-of-the-art method Dinomaly+, while maintaining stable performance as category cardinality increases.

cs.CV↗

arXiv2026

OpenCompass: A Universal Evaluation Platform for Large Language Models

In recent years, the field of artificial intelligence has undergone a paradigm shift from task-specific small-scale models to general-purpose large language models (LLMs). With the rapid iteration of LLMs, objective, quantitative, and comprehensive evaluation of their capabilities has become a critical link in advancing technological development. Currently, the mainstream static benchmark dataset-based evaluation methods face challenges such as the diversity of task types, inconsistent evaluation criteria, and fragmentation of data and processing workflows, making it difficult to efficiently conduct cross-domain and large-scale model evaluation. To address the aforementioned issues, this paper proposes and open-sources OpenCompass, a one-stop, scalable, and high-concurrency-supported general-purpose LLM evaluation platform. Adhering to the design philosophy of modularization and component decoupling, the platform boasts three core advantages: high compatibility, flexibility, and high concurrency. The core architecture of OpenCompass comprises five key components: the Configuration System, Task Partitioning Module, Execution and Scheduling Module, Task Execution Unit, and Result Visualization Module. Its workflow provides rule-based, LLM-as-a-Judge, and cascaded evaluators to adapt to the requirements of different task scenarios. Supporting mainstream benchmark datasets across multiple domains, including knowledge, reasoning, computation, science, language, code, etc., the platform offers a unified and efficient LLM evaluation tool for both academia and industry, facilitating the accurate identification of strengths and weaknesses of LLMs as well as their subsequent optimization.

cs.CL↗

arXiv2026

The superite phase and phase transition inducing multiscale solidification microstructures and segregations in steels

Based on classical concept, solidification of alloys is a direct transition from liquid phase to solid phase, by which dendrites and dendritic segregation are produced. Through in-situ and real time morphology observation and XRD test during solidification of three steels, a new superite phase featured as statistically oriented tiny structures was identified, and a general liquid-superite-solid phase transformation process is revealed. In the early solidification stage, the liquid alloys transit to dendrites composed of superite phase. Initiated from the boundaries of dendritic arms or dendrite grains, the superite phase transits to austenite grains within an initial dendritic arm, and expels solute elements to the residual superite phase. Mixed multi-phase microstructures are subsequently produced from the residual enriched superite phase. Here, although three steels exhibit different phase proportion and phase constitution in the superite-solid transition, they all follow above general transition mode. Multiscale microstructures and segregations are produced in the transition from superite to solid. These new findings change the basic understanding about the solidification of alloys, rediscover the formation mechanism on segregations and multiscale solidification microstructures, including dendrite pattern, solid dendritic arm, dendritic segregation, the mixed multi-phase microstructures, eutectic, inclusions and precipitate. These new findings are also crucial to the control of solidification microstructures and segregation in metals.

cond-mat.mtrl-sci↗

arXiv2026

Discover Fast Power Allocation Solution for Multi-Target Tracking via AlphaEvolve Evolution

Efficient radar resource allocation is a fundamental yet computationally challenging problem, as optimal solutions typically require iterative optimization with high complexity. Motivated by the need for real-time scheduling, robust generalization, and low data dependency, this paper proposes a novel paradigm that leverages large language model (LLM)-guided evolutionary search (AlphaEvolve) to autonomously discover a closed-form power allocation solution for multi-target tracking. The approach encodes high-dimensional radar states into physically inspired features, then evolves a compact and interpretable scoring function, which is transformed to feasible power allocations via a deterministic constraint-satisfying transformation. Extensive experiments demonstrate that the discovered closed-form solution achieves near-optimal tracking accuracy (average relative performance loss of only $1.51\%$), reliable generalization across diverse scenarios and target counts, and over three orders of magnitude speedup compared to conventional iterative solvers. These results highlight the potential of LLM-guided symbolic search to revolutionize not only radar resource management but also broader classes of engineering optimization problems.

eess.SP↗

arXiv2026

SAMoE-VLA: A Scene Adaptive Mixture-of-Experts Vision-Language-Action Model for Autonomous Driving

Recent advances in Vision-Language-Action (VLA) models have shown promising capabilities in autonomous driving by leveraging the understanding and reasoning strengths of Large Language Models(LLMs).However, our empirical analysis reveals that directly applying existing token-level MoE mechanisms--which are inherited from LLM architectures--to VLA models results in unstable performance and safety degradation in autonomous driving, highlighting a misalignment between token-based expert specialization and scene-level decision-making.To address this, we propose SAMoE-VLA, a scene-adaptive Vision-Language-Action framework that conditions expert selection on structured scene representations instead of token embeddings. Our key idea is to derive the MoE routing signal from bird's-eye-view (BEV) features that encapsulates traffic scene context, enabling scenario-dependent expert weighting and merging tailored to distinct driving conditions. Furthermore, to support temporally consistent reasoning across world-knowledge, perception, language, and action, we introduce a Conditional Cross-Modal Causal Attention mechanism that integrates world state, linguistic intent, and action history into a unified causal reasoning process. Extensive experiments on the nuScenes open loop planning dataset and LangAuto closed-loop benchmark demonstrate that SAMoE-VLA achieves state-of-the-art performance, outperforming prior VLA-based and world-model-based approaches with fewer parameters.Our code will be released soon.

cs.CV↗

arXiv2026

Root Cause Analysis Method Based on Large Language Models with Residual Connection Structures

Root cause localization remain challenging in complex and large-scale microservice architectures. The complex fault propagation among microservices and the high dimensionality of telemetry data, including metrics, logs, and traces, limit the effectiveness of existing root cause analysis (RCA) methods. In this paper, a residual-connection-based RCA method using large language model (LLM), named RC-LLM, is proposed. A residual-like hierarchical fusion structure is designed to integrate multi-source telemetry data, while the contextual reasoning capability of large language models is leveraged to model temporal and cross-microservice causal dependencies. Experimental results on CCF-AIOps microservice datasets demonstrate that RC-LLM achieves strong accuracy and efficiency in root cause analysis.

cs.AI↗

arXiv2025

New Constructions of Non-GRS MDS Codes, Recovery and Determination Algorithms for GRS Codes

In this paper, we propose a new method for constructing a class of non-GRS MDS codes. The lengths of these codes can reach up to $\frac{q+3}{2}$ (for finite fields of odd characteristic) and $\frac{q+4}{2}$ (for even characteristic), respectively. Owing to their special structure, we can use the Cauchy matrix method to obtain the necessary and sufficient conditions for these codes to be MDS codes and non-GRS MDS codes. Additionally, the inequivalence between these codes and twisted GRS codes is analyzed. Furthermore, we analyze the relationships among several existing classes of codes used for constructing non-GRS MDS codes, propose explicit constructions, and discuss the lengths of non-GRS MDS codes based on these constructions. Finally, we design two efficient algorithms to address two main problems in GRS code research, i.e., determining whether an unknown code $C$ is a GRS code from its generator matrix $G$, and recovering the key vectors $\bm{\alpha}$ and $\bm{v}$ such that $C = \GRS_{n,k}(\bm{\alpha}, \bm{v})$ if $C$ is indeed a GRS code. A computational complexity comparison of the proposed algorithms ($O(nk+n)$) with that of the Sidelnikov-Shestakov attack (exceeding $O(qk^2n+qk^3)$) shows that our methods offer superior computational efficiency.

cs.IT↗

arXiv2025

Rectifying LLM Thought from Lens of Optimization

Recent advancements in large language models (LLMs) have been driven by their emergent reasoning capabilities, particularly through long chain-of-thought (CoT) prompting, which enables thorough exploration and deliberation. Despite these advances, long-CoT LLMs often exhibit suboptimal reasoning behaviors, such as overthinking and excessively protracted reasoning chains, which can impair performance. In this paper, we analyze reasoning processes through an optimization lens, framing CoT as a gradient descent procedure where each reasoning step constitutes an update toward problem resolution. Building on this perspective, we introduce RePro (Rectifying Process-level Reward), a novel approach to refine LLM reasoning during post-training. RePro defines a surrogate objective function to assess the optimization process underlying CoT, utilizing a dual scoring mechanism to quantify its intensity and stability. These scores are aggregated into a composite process-level reward, seamlessly integrated into reinforcement learning with verifiable rewards (RLVR) pipelines to optimize LLMs. Extensive experiments across multiple reinforcement learning algorithms and diverse LLMs, evaluated on benchmarks spanning mathematics, science, and coding, demonstrate that RePro consistently enhances reasoning performance and mitigates suboptimal reasoning behaviors.

cs.CL↗

arXiv2025

Two-Step Decoding of Binary $2\times2$ Sum-Rank-Metric Codes

We address an open problem posed by Chen-Cheng-Qi (IEEE Trans.\ Inf.\ Theory, 2025): can the decoding of binary sum-rank-metric codes $\SR(C_1,C_2)$ with $2\times2$ matrix blocks be reduced entirely to decoding the constituent Hamming-metric codes $C_1$ and $C_2$ without the additional requirement $d_1\ge\tfrac{2}{3}d_{\mathrm{sr}}$ used in their fast decoder? We answer this in the affirmative by exhibiting a simple two-step procedure: first uniquely decode $C_2$, then apply a single error-erasure decoding for $C_1$. This shows that the restrictive hypothesis $d_1\ge\tfrac{2}{3}d_{\mathrm{sr}}$ is theoretically unnecessary. The resulting decoder achieves unique decoding up to $\lfloor (d_{\mathrm{sr}}-1)/2\rfloor$ with overall cost $T_2+T_1$, where $T_2$ and $T_1$ are the complexities of the Hamming decoders for $C_2$ and $C_1$, respectively. We further show that this reduction is asymptotically optimal in a black-box model, as any sum-rank decoder must inherently decode the constituent Hamming codes. For BCH or Goppa instantiations over $\F_4$, the decoder runs in $O(\ell^2)$ time.

cs.IT↗

arXiv2025

ATLAS: A High-Difficulty, Multidisciplinary Benchmark for Frontier Scientific Reasoning

The rapid advancement of Large Language Models (LLMs) has led to performance saturation on many established benchmarks, questioning their ability to distinguish frontier models. Concurrently, existing high-difficulty benchmarks often suffer from narrow disciplinary focus, oversimplified answer formats, and vulnerability to data contamination, creating a fidelity gap with real-world scientific inquiry. To address these challenges, we introduce ATLAS (AGI-Oriented Testbed for Logical Application in Science), a large-scale, high-difficulty, and cross-disciplinary evaluation suite composed of approximately 800 original problems. Developed by domain experts (PhD-level and above), ATLAS spans seven core scientific fields: mathematics, physics, chemistry, biology, computer science, earth science, and materials science. Its key features include: (1) High Originality and Contamination Resistance, with all questions newly created or substantially adapted to prevent test data leakage; (2) Cross-Disciplinary Focus, designed to assess models' ability to integrate knowledge and reason across scientific domains; (3) High-Fidelity Answers, prioritizing complex, open-ended answers involving multi-step reasoning and LaTeX-formatted expressions over simple multiple-choice questions; and (4) Rigorous Quality Control, employing a multi-stage process of expert peer review and adversarial testing to ensure question difficulty, scientific value, and correctness. We also propose a robust evaluation paradigm using a panel of LLM judges for automated, nuanced assessment of complex answers. Preliminary results on leading models demonstrate ATLAS's effectiveness in differentiating their advanced scientific reasoning capabilities. We plan to develop ATLAS into a long-term, open, community-driven platform to provide a reliable "ruler" for progress toward Artificial General Intelligence.

cs.CL↗

arXiv2025

Improved AntiGriesmer Bounds for Linear Anticodes and Applications

This paper improves the antiGriesmer bound for linear anticodes previously established by Chen and Xie (Journal of Algebra, 673 (2025) 304-320). While the original bound required the code length to satisfy $n < q^{k-1}$ and the dual code to have minimum distance at least 3, our main result removes the length restriction and relaxes the dual distance condition to at least 2. Specifically, we prove that for any $[n,k]_q$ linear anticode $\mathcal{C}$ over $\mathbb{F}_q$ with diameter $\delta$ and $d(\mathcal{C}^\perp) \geq 2$, the inequality \[ n \leq \sum_{i=0}^{k-1} \left\lfloor \frac{\delta}{q^i} \right\rfloor \] holds. This generalization significantly broadens the applicability of the antiGriesmer bound. We derive several corollaries, including lower bounds on the diameter $\delta$ in terms of $n$ and $k$, upper bounds on the code length $n$, and constraints on the dimension $k$. Applications to the construction and classification of linear codes with few weights are also discussed, along with examples demonstrating that our new bound can be sharper than previous ones. Our work unifies and extends earlier findings, providing a more comprehensive framework for studying linear anticodes and their properties.

cs.IT↗

arXiv2025

Improved Decoding Algorithms for MDS and Almost-MDS Codesfrom Twisted GRS Codes

In this paper, firstly, we study decoding of a general class of twisted generalized Reed-Solomon (TGRS) codes and provide a precise characterization of the key equation for TGRS codes and propose a decoding algorithm. Secondly, we further study decoding of almost-MDS TGRS codes and provide a decoding algorithm. These two decoding algorithms are more efficient in terms of performance compared with the decoding algorithms presented in [Sun et al., IEEE-TIT, 2024] and [Sui et al., IEEE-TIT, 2023] respectively.

cs.IT↗

arXiv2025

A Multi-Level Framework for Multi-Objective Hypergraph Partitioning: Combining Minimum Spanning Tree and Proximal Gradient

This paper proposes an efficient hypergraph partitioning framework based on a novel multi-objective non-convex constrained relaxation model. A modified accelerated proximal gradient algorithm is employed to generate diverse $k$-dimensional vertex features to avoid local optima and enhance partition quality. Two MST-based strategies are designed for different data scales: for small-scale data, the Prim algorithm constructs a minimum spanning tree followed by pruning and clustering; for large-scale data, a subset of representative nodes is selected to build a smaller MST, while the remaining nodes are assigned accordingly to reduce complexity. To further improve partitioning results, refinement strategies including greedy migration, swapping, and recursive MST-based clustering are introduced for partitions. Experimental results on public benchmark sets demonstrate that the proposed algorithm achieves reductions in cut size of approximately 2\%--5\% on average compared to KaHyPar in 2, 3, and 4-way partitioning, with improvements of up to 35\% on specific instances. Particularly on weighted vertex sets, our algorithm outperforms state-of-the-art partitioners including KaHyPar, hMetis, Mt-KaHyPar, and K-SpecPart, highlighting its superior partitioning quality and competitiveness. Furthermore, the proposed refinement strategy improves hMetis partitions by up to 16\%. A comprehensive evaluation based on virtual instance methodology and parameter sensitivity analysis validates the algorithm's competitiveness and characterizes its performance trade-offs.

cs.LG↗

arXiv2025

New Bounds for Linear Codes with Applications

Bounds on linear codes play a central role in coding theory, as they capture the fundamental trade-off between error-correction capability (minimum distance) and information rate (dimension relative to length). Classical results characterize this trade-off solely in terms of the parameters $n$, $k$, $d$ and $q$. In this work we derive new bounds under the additional assumption that the code contains a nonzero codeword of weight $w$.By combining residual-code techniques with classical results such as the Singleton and Griesmer bounds,we obtain explicit inequalities linking $n$, $k$, $d$, $q$ and $w$. These bounds impose sharper restrictions on admissible codeword weights, particularly those close to the minimum distance or to the code length. Applications include refined constraints on the weights of MDS codes, numerical restrictions on general linear codes, and excluded weight ranges in the weight distribution. Numerical comparisons across standard parameter sets demonstrate that these $w$-aware bounds strictly enlarge known excluded weight ranges and sharpen structural limitations on linear codes.

cs.IT↗

arXiv2025

Asymptotic Properties of a Forward-Backward-Forward Differential Equation and Its Discrete Version for Solving Quasimonotone Variational Inequalities

This paper investigates the asymptotic behavior of a forward-backward-forward (FBF) type differential equation and its discrete counterpart for solving quasimonotone variational inequalities (VIs). Building on recent continuous-time dynamical system frameworks for VIs, we extend these methods to accommodate quasimonotone operators. We establish weak and strong convergence under significantly relaxed conditions, without requiring strong pseudomonotonicity or sequential weak-to-weak continuity. Additionally, we prove ergodic convergence of the continuous trajectories, offering further insight into the long-term stability of the system. In the discrete setting, we propose a novel Bregman-type algorithm that incorporates a nonmonotone adaptive step-size rule based on the golden ratio technique. A key contribution of this work is demonstrating that the proposed method ensures strong convergence under the assumption of uniform continuity of the operator, thereby relaxing the standard Lipschitz continuity requirement prevalent in existing methods. Numerical experiments, including infinite-dimensional and non-Lipschitz cases, are presented to illustrate the improved convergence and broader applicability of the proposed approach.

math.OC↗

arXiv2025

When Tools Hurt LLM Reasoning: State-Dependent Belief Revision under External Evidence

Tool use is often assumed to monotonically improve reasoning, where external evidence is expected to help when relevant and be ignored when irrelevant. We show that this assumption fails in a state-dependent way. Across benchmarks with Python and Wikipedia tools, external evidence reliably helps when initial beliefs are weak, but can flip already-correct answers when those beliefs are strong. We frame this as a misallocation of revision authority, arguing that deferring to external evidence is suboptimal when internal support for the correct answer surpasses the tool's expected output quality. This predicts that harm should concentrate on high-confidence no-tool cases. We test this prediction with threshold localization, wrong-trace audits, and a same-clue intervention showing that revision framing changes the damage caused by misleading evidence. These findings suggest that mixed no-tool/tool-assisted inference should arbitrate authority rather than privilege tool evidence by default. As a minimal demonstration, we introduce CASE, a label-free controller that selects between no-tool and tool-assisted trajectories using answer-state certainty and improves over existing confidence-based arbitration baselines.The code for our experiments is available at https://github.com/epsilondylan/State-Dependent-Belief-Revision.

cs.CL↗