cs papers | Gist.Science

PointCloud-ID: Deterministic Cross-Format Identity and Attribute-Aware Integrity Verification for Geospatial Point Clouds

This paper presents PointCloud-ID, a deterministic framework that establishes stable identity and enables spatially localized integrity verification for geospatial point clouds by canonicalizing data under versioned profiles and constructing attribute-partitioned Merkle trees, thereby distinguishing content-preserving re-encodings from actual modifications.

A. Urias2026-08-12

💻 computer science

A Lightweight Diffusion-Augmented Framework for Efficient Children Facial Expression Recognition

This paper proposes a lightweight diffusion-augmented framework that combines a lightweight CNN, diffusion-based data augmentation, and knowledge distillation to achieve high-performance, real-time facial expression recognition in children on resource-constrained devices, outperforming existing baselines on LIRIS-CSE and CAFE datasets.

Nitin Arora2026-08-12

💻 computer science

A Hybrid Graph-Based and AI-Driven Cooperative System for Overtaking Collision Avoidance

This paper presents IG-COAS, a hybrid cooperative system that enhances overtaking safety on two-lane roads by integrating V2X communication, dynamic graph-based environment modeling, and cloud-assisted machine learning to provide real-time, predictive collision risk assessment and decision support.

Safa Batita, Achraf Makni, Ikram Amous2026-08-12

💻 computer science

A Two-Level LSTM Framework for Traffic Forecasting and Early Congestion Prediction in SD-WAN Networks

This paper proposes a two-level LSTM framework for SD-WAN networks that integrates channel load prediction and critical congestion classification to enable proactive route management, demonstrating superior forecasting accuracy and decision reliability compared to traditional methods.

Vitalii Savchenko, Yurii Kravchenko, Serhii Stavytskyi, Olga Leshchenko, Olena Starkova2026-08-12

💻 computer science

Cross-Version Domain Adaptation and Open-Set Attack Screening for Transferable Wi-Fi Intrusion Detection

This paper proposes a cross-version domain adaptation framework that combines IEEE 802.11 feature harmonization, DANN/CORAL alignment, and confidence-based open-set screening to achieve high-accuracy, real-time Wi-Fi intrusion detection across shifting environments while effectively identifying previously unseen attack families with minimal target-domain labeling.

Hadeer Hassan Ibrahim, Rasha M. Ismail, Tamer AbdelKader, Alshaimaa Abo-Alian2026-08-12

💻 computer science

PPSNet: Perceptual Prior Similarity-guided Network for Class-agnostic Counting

This paper introduces PPSNet, a novel class-agnostic counting framework that theoretically validates the Extract-and-Match paradigm and addresses the limitations of bounding box-based visual exemplars by employing a Perceptual Prior Construction Module to refine scale distribution and a Similarity-guided Weighting Module to enhance feature matching, thereby achieving superior performance across visual, text, and zero-shot prompting scenarios.

Shengwei Jia, Junhui Liu, Jiahao Wang, Yongqiang Cui, Shihui Zhang2026-08-12

💻 computer science

A Physics-Informed MobileNetV2 Framework for Thermal Infrared Gas Classification: Combining Thermodynamic Input Features and Loss-Level Physics Penalization

This paper introduces a Physics-Informed MobileNetV2 framework for thermal infrared gas classification that integrates thermodynamic input features and physics-penalized loss functions to achieve high accuracy and zero false alarms for mixture detection, offering a favorable trade-off between raw performance and physical plausibility for safety-critical applications.

Kayode Oladele Olumurewa, Oluwadara O. Adeniji, Johnson Bisi Oluwagbemi2026-08-12

💻 computer science

Intelligent Prompt Filtering and Jailbreak Detection for Safe Deployment of Generative

This paper proposes an Adaptive Multi-Dataset Framework (AMDF) that integrates three benchmark datasets with Sentence-BERT and XGBoost to effectively detect and filter prompt injection and jailbreak attacks, achieving 84% accuracy and enhancing the generalization of LLM defenses against evolving threats.

Anuradha Singh, Aarti Singh, Nikki Singh, Vidhi Singh⁴2026-08-12

💻 computer science

Breaking Penalty Plateaus with Quantum-Inspired Improved Particle Swarm Optimization

This paper proposes a quantum-inspired improved particle swarm optimization (QI-PSO) that replaces classical velocity-driven movement with bounded-potential quantum displacement laws to effectively overcome penalty plateaus in constrained multimodal optimization, demonstrating significant error reductions on difficult benchmarks while retaining the classical approach's superiority on exploitation-dominant problems.

Prashant Pandey, Raju Prajapati2026-08-12

💻 computer science

Human-Centered Cloud Automated Provisioning with Deep Reinforcement Learning for Adaptive Computing

This paper empirically demonstrates the limitations of static cloud provisioning methods in handling dynamic workloads and SLA constraints using Google Cluster traces, thereby advocating for the adoption of Deep Reinforcement Learning as a superior, adaptive, and human-centered approach for automated resource orchestration.

Kavita Srivastava, Manisha Agarwal2026-08-12

← Previous Next →