cs papers | Gist.Science

Diversity-Based Fitness Regularization in Genetic Algorithms: A Methodological Audit Across Population Sizes

This paper audits a diversity-based fitness regularization method in genetic algorithms against a magnitude-matched noise control protocol, finding that its purported benefits are largely indistinguishable from unstructured noise and driven by outliers, thereby supporting the method only in a narrow regime while establishing a rigorous framework for future evaluations of inertia mechanisms.

Tilan Ukwatta2026-09-01

💻 computer science

DMASNet: A Defect Detection Network for PCBs via Multi-Scale Feature Collaborative Optimization

This paper introduces DMASNet, a lightweight and efficient defect detection framework for printed circuit boards that leverages multi-scale feature optimization, dynamic upsampling, and specialized loss functions to significantly improve accuracy and real-time inference speed on industrial hardware.

Yingying Zhang, Yu Zhao, Tonghua Wang, Ya Gao, Zhiwei Chen2026-08-31

💻 computer science

Quantum Local Density of States for Random k-SAT: An Amplitude-Estimation Primitive and a Clause-Width Regime for Quantum Advantage

This paper introduces a quantum Local Density of States (LDOS) primitive for random k-SAT that uses amplitude estimation to efficiently estimate the residual satisfying fraction, demonstrating a quantum advantage for clause widths of four or higher while clarifying that the positivity fraction is primarily a structural counting effect rather than a signal of the freezing transition.

Michail Gerogiannis, Dimitris Ntalaperas, Nikos Konofaos2026-08-31

💻 computer science

Adaptive Reconstruction-Aware Evidence Fusion for Generalizable Diffusion-Generated Image Detection

This paper proposes an adaptive reconstruction-aware evidence fusion framework that integrates spatial discriminative and reconstruction-aware branches through anomaly-guided interaction and reliability routing to achieve robust, generalizable detection of diffusion-generated images across unseen generators and post-processing perturbations.

Song Shen, Tingnian He, Shenghui Ji, Xiaolin Wei2026-08-31

💻 computer science

Calibrated Residual Modeling for Explainable Anomaly Detection in Multivariate Sensor Streams

This paper proposes a hybrid deep statistical framework that integrates a temporal convolutional autoencoder, a probabilistic residual model, and extreme value calibration to achieve robust, interpretable, and adaptive anomaly detection in multivariate sensor streams, demonstrating superior performance and reduced detection delays across six diverse benchmarks compared to transformer and contrastive baselines.

Rui Shi2026-08-31

💻 computer science

Research on the Generation of Tibetan Word-levelAdversarial Samples Based on Morphosyntactic Linkage

This paper proposes TWLASBMSL, a novel method for generating Tibetan word-level adversarial samples that integrates morphosyntactic linkage to ensure grammatical correctness while effectively deceiving target models in sentiment analysis and text classification tasks.

Btsan lhaMgon, Qun Nuo, Nyima Tashi2026-08-31

💻 computer science

DRS-Track: An Enhanced Detection-Driven Framework for Multi- Object Tracking

This paper introduces DRS-Track, an enhanced multi-object tracking framework that integrates dynamic sampling, receptive-field attention, and a cross-scale shared detection head to overcome challenges like scale variations and occlusions, achieving superior performance on both a self-constructed dataset and the KITTI benchmark.

Wei Yang, Qi Ji, Chengyi Hou, Mushuo Liu, Jianqiang Gong2026-08-31

💻 computer science

Integrating Neuromorphic Computing and Clinical Biomarkers for Enhanced Liver Disease Detection

This paper proposes an energy-efficient, robust, and interpretable neuromorphic computing system that utilizes spiking neural networks and hardware-aware training to achieve high-accuracy, real-time liver disease detection using clinical biomarkers, offering a superior alternative to conventional computationally intensive models.

Ashish V Saywan, Soni A Chaturvedi2026-08-31

💻 computer science

Digital Twin Feedback for Predictive Maintenance in Industrial IoT Environments

The paper proposes the Hierarchical Sensor-Fused Edge Intelligence with Digital Twin Feedback (HSEI-DTF) framework, a multi-layered predictive maintenance system that integrates edge-based multi-modal signal processing, hybrid machine learning models, and physics-informed digital twin synchronization to significantly reduce equipment downtime and false-negative rates while achieving high accuracy and low latency in industrial IoT environments.

Megha Patil, Yogesh Bhirud, Dhanashree Barbole2026-08-31

💻 computer science

Beyond Pairwise Relations: Dynamic Manipulation Hypergraphs for Vision-Based Human Activity Recognition

This paper proposes a dynamic manipulation hypergraph framework that models evolving multi-entity interactions as higher-order relational units rather than pairwise edges, achieving significant performance improvements in fine-grained human activity recognition on EPIC-KITCHENS-100/VISOR and Assembly101 datasets.

Fatemeh Ziaeetabar2026-08-31

← Previous Next →