arXiv subjects

Feng Cao

Publications and source records attributed to Feng Cao.

12 recordsLinked to original sources

arXiv2026

ArcAD: Anomaly-Rectified Calibration for Cold-Start Supervised Anomaly Detection

The deployment of Industrial Anomaly Detection (IAD) in real-world manufacturing frequently encounters a challenging cold-start bottleneck, in which limited normal samples fail to represent the full normal distribution and only a few anomalies are available. Under such a regime, existing methods struggle to form compact normal boundaries and fail to effectively exploit supervised signals from rare defects. To address this challenge, we propose Anomaly-Rectified Cold-start AD (ArcAD), a plug-and-play calibration framework for reconstruction-based IAD baselines. ArcAD follows a push-pull learning paradigm to construct a compact and discriminative normal boundary under data scarcity. On the one hand, ArcAD projects limited normal samples onto a hypersphere and pulls them into multiple compact clusters to maximize coverage of the normal manifold. On the other hand, it synthesizes pseudo-anomalies on the hypersphere and leverages real anomalies to push the boundary inward and sharpen anomaly discrimination. Extensive experiments on MVTec-AD, VisA, Real-IAD, and MANTA demonstrate that ArcAD significantly outperforms state-of-the-art supervised and unsupervised methods in both single-class and multi-class settings under cold-start conditions. Code is available at: https://github.com/LGC-AD/ArcAD.

cs.CV↗

arXiv2025

Multidimensional Rubric-oriented Reward Model Learning via Geometric Projection Reference Constraints

The integration of large language models (LLMs) into medical practice offers transformative potential, yet their real-world clinical applicability remains constrained by critical alignment issues: (1) a misalignment between static evaluation benchmarks and the dynamic cognitive demands of clinical practice, (2) challenges in adapting to continuously evolving, multi-source medical standards, and (3) the limited capacity of conventional reward models to reflect nuanced, multi-dimensional medical quality criteria. To overcome these limitations, we introduce MR-RML (Multidimensional Rubric-oriented Reward Model Learning) with GPRC (Geometric Projection Reference Constraints)-a novel alignment framework that structured medical standards into a multi-perspective matrix to guide both data generation and model optimization. Our approach introduces three key innovations: (1) a medical standard system that embeds domain-specific guidelines throughout the training pipeline; (2) an independent multi-dimensional reward model that decomposes evaluation criteria, transitioning from rule-based or LLM-based scoring to internalized reward modeling for better evaluation performance; and (3) geometric projection reference constraints that translate clinical cognitive logic into mathematical regularization, aligning scoring gradients with clinical reasoning and facilitating training with synthetically generated data. Extensive evaluations on the authoritative medical benchmark Healthbench demonstrate that our method significantly boosts the performance of the base Qwen-32B model, with improvements of 45% on the full subset and 85% on the hard subset. It achieves state-of-the-art results among open-source LLMs, scoring 62.7 (full) and 44.7 (hard), while also surpassing the majority of closed-source models.

cs.AI↗

arXiv2025

Extended Triangular Method: A Generalized Algorithm for Contradiction Separation Based Automated Deduction

Automated deduction lies at the core of Artificial Intelligence (AI), underpinning theorem proving, formal verification, and logical reasoning. Despite decades of progress, reconciling deductive completeness with computational efficiency remains an enduring challenge. Traditional reasoning calculi, grounded in binary resolution, restrict inference to pairwise clause interactions and thereby limit deductive synergy among multiple clauses. The Contradiction Separation Extension (CSE) framework, introduced in 2018, proposed a dynamic multi-clause reasoning theory that redefined logical inference as a process of contradiction separation rather than sequential resolution. While that work established the theoretical foundation, its algorithmic realization remained unformalized and unpublished. This work presents the Extended Triangular Method (ETM), a generalized contradiction-construction algorithm that formalizes and extends the internal mechanisms of contradiction separation. The ETM unifies multiple contradiction-building strategies, including the earlier Standard Extension method, within a triangular geometric framework that supports flexible clause interaction and dynamic synergy. ETM serves as the algorithmic core of several high-performance theorem provers, CSE, CSE-E, CSI-E, and CSI-Enig, whose competitive results in standard first-order benchmarks (TPTP problem sets and CASC 2018-2015) empirically validate the effectiveness and generality of the proposed approach. By bridging theoretical abstraction and operational implementation, ETM advances the contradiction separation paradigm into a generalized, scalable, and practically competitive model for automated reasoning, offering new directions for future research in logical inference and theorem proving.

cs.AI↗

arXiv2025

HuiduRep: A Robust Self-Supervised Framework for Learning Neural Representations from Extracellular Recordings

Extracellular recordings are transient voltage fluctuations in the vicinity of neurons, serving as a fundamental modality in neuroscience for decoding brain activity at single-neuron resolution. Spike sorting, the process of attributing each detected spike to its corresponding neuron, is a pivotal step in brain sensing pipelines. However, it remains challenging under low signal-to-noise ratio (SNR), electrode drift and cross-session variability. In this paper, we propose HuiduRep, a robust self-supervised representation learning framework that extracts discriminative and generalizable features from extracellular recordings. By integrating contrastive learning with a denoising autoencoder, HuiduRep learns latent representations that are robust to noise and drift. With HuiduRep, we develop a spike sorting pipeline that clusters spike representations without ground truth labels. Experiments on hybrid and real-world datasets demonstrate that HuiduRep achieves strong robustness. Furthermore, the pipeline outperforms state-of-the-art tools such as KiloSort4 and MountainSort5. These findings demonstrate the potential of self-supervised spike representation learning as a foundational tool for robust and generalizable processing of extracellular recordings. Code is available at: https://github.com/IgarashiAkatuki/HuiduRep

eess.SP↗

arXiv2024

CNMBERT: A Model for Converting Hanyu Pinyin Abbreviations to Chinese Characters

The task of converting Hanyu Pinyin abbreviations to Chinese characters is a significant branch within the domain of Chinese Spelling Correction (CSC). It plays an important role in many downstream applications such as named entity recognition and sentiment analysis. This task typically involves text-length alignment and seems easy to solve; however, due to the limited information content in pinyin abbreviations, achieving accurate conversion is challenging. In this paper, we treat this as a fill-mask task and propose CNMBERT, which stands for zh-CN Pinyin Multi-mask BERT Model, as a solution to this issue. By introducing a multi-mask strategy and Mixture of Experts (MoE) layers, CNMBERT outperforms fine-tuned GPT models and ChatGPT-4o with a 61.53% MRR score and 51.86% accuracy on a 10,373-sample test dataset.

cs.CL↗

arXiv2021

Insect-Computer Hybrid System for Autonomous Search and Rescue Mission

There is still a long way to go before artificial mini robots are really used for search and rescue missions in disaster-hit areas due to hindrance in power consumption, computation load of the locomotion, and obstacle-avoidance system. Insect-computer hybrid system, which is the fusion of living insect platform and microcontroller, emerges as an alternative solution. This study demonstrates the first-ever insect-computer hybrid system conceived for search and rescue missions, which is capable of autonomous navigation and human presence detection in an unstructured environment. Customized navigation control algorithm utilizing the insect's intrinsic navigation capability achieved exploration and negotiation of complex terrains. On-board high-accuracy human presence detection using infrared camera was achieved with a custom machine learning model. Low power consumption suggests system suitability for hour-long operations and its potential for realization in real-life missions.

cs.RO↗

arXiv2019

Transition fronts of two species competition lattice systems in random media

The current paper is devoted to the study of existence and non-existence of transition fronts for two species competition lattice system in random media, and explore the influence of randomness of the media on the wave profiles and wave speeds of such transition fronts. We first establish comparison principle for sub-solutions and super-solutions of the related cooperative system. Next, under some proper assumptions, we construct appropriate sub-solutions and super-solutions for the cooperative system. Finally, we show that random transition fronts exist if their least mean speed is greater than an explicit threshold and there is no random transition front with least mean speed less than the threshold.

math.DS↗

arXiv2019

Transition fronts of KPP-type lattice random equations

In this paper, we investigate the existence and stability of random transition fronts of KPP-type lattice equations in random media, and explore the influence of the media and randomness on the wave profiles and wave speeds of such solutions. We first establish comparison principle for sub-solutions and super-solutions of KPP type lattice random equations and prove the stability of positive constant equilibrium solution. Next, by constructing appropriate sub-solutions and super-solutions, we show the existence of random transition fronts. Finally, we prove the stability of random transition fronts of KPP-type lattice random equations.

math.DS↗

arXiv2019

Stability and uniqueness of generalized traveling waves of lattice Fisher-KPP equations in heterogeneous media

In this paper, we investigate the stability and uniqueness of generalized traveling wave solutions of lattice Fisher-KPP equations with general time and space dependence. We first show the existence, uniqueness, and stability of strictly positive entire solutions of such equations. Next, we show the stability and uniqueness of generalized traveling waves connecting the unique strictly positive entire solution and the trivial solution zero. Applying the general stability and uniqueness theorem, we then prove the existence, stability and uniqueness of periodic traveling wave solutions of lattice Fisher-KPP equations in time and space periodic media, and the existence, stability and uniqueness of generalized traveling wave solutions of lattice Fisher-KPP equations in time heterogeneous media. The general stability result established in this paper implies that the generalized traveling waves obtained in many cases are asymptotically stable under well-fitted perturbation.

math.DS↗

arXiv2017

Spreading speeds and transition fronts of lattice KPP equations in time heterogeneous media

The current paper is devoted to the study of spreading speeds and transition fronts of lattice KPP equations in time heterogeneous media. We first prove the existence, uniqueness, and stability of spatially homogeneous entire positive solutions. Next, we establish lower and upper bounds of the (generalized) spreading speed intervals. Then, by constructing appropriate sub-solutions and super-solutions, we show the existence and continuity of transition fronts with given front position functions. Also, we prove the existence of some kind of critical front.

math.DS↗

arXiv2013

Convergence in Comparable Almost Periodic Reaction-Diffusion Systems with Dirichlet Boundary Condition

The paper is to study the asymptotic dynamics in nonmonotone comparable almost periodic reaction-diffusion system with Dirichlet boundary condition, which is comparable with uniformly stable strongly order-preserving system. By appealing to the theory of skew-product semiflows, we obtain the asymptotic almost periodicity of uniformly stable solutions to the comparable reaction-diffusion system.

math.DS↗

arXiv2012

Group Actions on Monotone Skew-Product Semiflows with Applications

We discuss a general framework of monotone skew-product semiflows under a connected group action. In a prior work, a compact connected group $G$-action has been considered on a strongly monotone skew-product semiflow. Here we relax the requirement of strong monotonicity of the skew-product semiflows and the compactness of $G$, and establish a theory concerning symmetry or monotonicity properties of uniformly stable 1-cover minimal sets. We then apply this theory to show rotational symmetry of certain stable entire solutions for a class of non-autonomous reaction-diffusion equations on $\RR^n$, as well as monotonicity of stable travelling waves of some nonlinear diffusion equations in time recurrent structures including almost periodicity and almost automorphy.

math.DS↗