💻 Category

cs

2608 papers

💻 computer science

When Modality Importance Does Not Translate into Capacity Gains: A Controlled Study of Asymmetric Multimodal Recommendation

This controlled study demonstrates that increasing representational capacity for the more informative text modality in FREEDOM-style multimodal recommenders fails to yield consistent accuracy gains because the added capacity lacks a direct gradient path to the primary ranking objective, highlighting that modality informativeness does not automatically translate into performance improvements through asymmetric architectural allocation.

Emin Talip Demirkiran2026-09-21
💻 computer science

Towards Quantum Decision Intelligence: A Four-Qubit Architecture for Venture Capital's Non-Linear Dilemmas on a 156-Qubit IBM Heron Processor

This paper presents HQDIS, a four-qubit variational quantum circuit tested on IBM's 156-qubit Heron processor for venture capital screening, which demonstrates high hardware-simulation fidelity and performance comparable to classical models on a small synthetic benchmark, though it is outperformed by classical algorithms on a second synthetic dataset, highlighting the current limitations of quantum advantage in this domain.

Enrique Díaz de León López2026-09-21
💻 computer science

Exact Incremental Updates for Continual Sequential Recommendation

This paper demonstrates that while a closed-form temporal linear model cannot match the accuracy of neural baselines like CSTRec in continual sequential recommendation, its sufficient-statistics incremental update strategy offers a numerically exact and computationally efficient alternative to full re-solving, whereas Woodbury-based updates fail due to memory constraints when update blocks exceed the item catalog size.

Emin Talip Demirkiran2026-09-21
💻 computer science

Cyber-QEE Under Rigorous Validation: Ablation, Missingness, and Near-Duplicate Robustness in Network Intrusion Detection

This study demonstrates that while the Cyber-QEE stochastic energy representation can conditionally improve XGBoost-based intrusion detection under specific severe missingness scenarios, it does not provide consistent, independent predictive value over standard features and often performs comparably to permuted or noise controls, suggesting its benefits are methodological rather than universally generalizable.

Hussein Dedy2026-09-21
💻 computer science

History-First Reliability-Gated Fusion for Sampled-Peak System and GPU Memory Utilization Prediction in HPC Clusters

This paper presents a history-first, reliability-gated fusion framework that combines exact-script historical caching with a LoRA-adapted Qwen2.5-Coder encoder to significantly improve the prediction accuracy of system and GPU memory utilization in HPC clusters while reducing computational overhead through selective inference.

Pan Chang, Xueru Chen, Guangchao Hu2026-09-21
💻 computer science

Deep Transfer Learning for Automated Cervical Cell-Type Classification in Pap-Smear Images: A Comparative and Explainable Study

This study demonstrates that a ResNet50-based deep transfer learning model, enhanced with explainable AI techniques like Grad-CAM, achieves high accuracy (91.50%) in classifying five distinct cervical cell types from Pap-smear images, though it faces specific challenges in distinguishing between Koilocytotic and Metaplastic classes.

Prashansa D. Choksi, Vipul B. Bambhaniya, Chetan Shingadiya2026-09-21
💻 computer science

Artificial Intelligence and Spirituality: A Systematic Scoping Review of Human-AI Engagement

This systematic scoping review of 19 studies published between late 2022 and 2026 reveals that conversational AI is increasingly utilized for diverse spiritual and existential purposes, serving as a unique source of authority, emotional disclosure, and ritual facilitation, while simultaneously raising unresolved ethical concerns regarding bias, privacy, and the nature of spiritual authenticity.

Anna Burcombe, Patricia Carlisle, Paul Best2026-09-21
💻 computer science

Personalized Federated Learning Under Severe Statistical Heterogeneity: A Multi-Dataset Analysis of Accuracy, Tail Performance, and Client Fairness

This paper introduces a rigorous, client-centric multi-dataset evaluation framework that analyzes personalized federated learning under severe statistical heterogeneity by jointly assessing global accuracy, lower-tail performance, and fairness metrics to determine whether personalization truly benefits the most poorly served clients.

Md Shahanur Islam Shagor2026-09-21
💻 computer science

CONFLUENCE: Does Cross-Scale Graph Attention Help Watershed Vegetation Forecasting?

This paper challenges the assumption that graph neural networks improve watershed vegetation forecasting by demonstrating that, in the Potomac River case study, cross-scale graph attention and even physically derived topologies often underperform simple baselines or random connections, highlighting the critical need for rigorous, matched-capacity controls to avoid false positives in ecological modeling.

Andrew Ma2026-09-21
💻 computer science

Auditing Level-Capacity Ablations in Single-Stage Object Detectors: A Decoupling Patch, a Fixed-Calibre Repair, and the Limits of Budget-Mismatch Ratios

This paper demonstrates that the Level-Budget Mismatch Ratio (LBMR), a metric proposed to guide capacity reallocation in single-stage object detectors, fails as an actionable optimization tool due to non-linear accuracy responses, coupled architectural dependencies, and Pareto-dominated outcomes, ultimately showing that a decoupling patch and fixed-calibre repair offer no measurable benefit over default configurations.

Pinchao Huang, Xingying Cai2026-09-21