cs papers | Gist.Science

Benchmarking Gradient Boosting and Explainable AI for Credit Default Prediction on Imbalanced Financial Data: A Leakage-Safe Multi-Seed Evaluation with SHAP and LIME

This paper establishes a rigorous, leakage-safe benchmarking protocol for credit default prediction that demonstrates gradient boosting ensembles significantly outperform logistic regression and TabNet on imbalanced financial data, while revealing that post-hoc threshold tuning renders synthetic resampling redundant and highlighting the critical need to validate explanation discordance between SHAP and LIME for model risk management.

Manpreet Singh, Rohith Reddy Bellibatlu, Yash Jajoo, Muhammad Zeeshan, Rathan Ramachandra, Akshatha Srikantha, Rahul Jos (…)2026-09-03

💻 computer science

Beyond Prompt Injection: Trust-Boundary Security Assurance for LLM-Integrated and Agentic Applications

This paper proposes a methodological trust-boundary security assurance framework that models LLM-integrated and agentic systems to translate evolving semantic threats into verifiable, testable security controls and evidence-based reporting.

Nazar Waheed2026-09-03

💻 computer science

Automated Disease Prediction and Prescription System with Regional Kannada Synonym Integration

This paper presents an automated disease prediction and prescription system that utilizes an RNN-LSTM model to accurately translate regional Kannada medical synonyms into English, thereby outperforming existing tools like Google Translate and Whisper in context recognition to assist healthcare professionals in managing high patient volumes.

Giridhara Gouda, Suma R2026-09-03

💻 computer science

Evidence Quality in Assurance-Oriented Benchmark Construction

This paper characterizes the evidence quality of public software incidents for assurance-oriented benchmark construction by analyzing a 40-case dataset to reveal significant gaps in reconstructability, pairing, and source provenance, ultimately releasing the AIRR-40 registry to support more rigorous, task-relative benchmarking.

Byungsik Seo2026-09-03

💻 computer science

biotrace2csv: Data Extraction Tool for BioTrace+ PDF Data Reports

biotrace2csv is a free, open-source command-line tool that automates the extraction of heart rate variability and physiological data from BioTrace+ PDF reports into CSV format, achieving 100% accuracy with manual extraction while reducing processing time by approximately 50-fold.

Metehan Özsoy2026-09-03

💻 computer science

Finite Evidence, Infinite Horizons: The Artificial Age Score as a Case Study in Longitudinal AI Measurement

This paper argues that finite longitudinal AI evaluation records, exemplified by the Artificial Age Score (AAS), cannot determine infinite-horizon system properties or convergence without explicit stipulated continuation models, as demonstrated through counterexamples and stress tests showing that bounded observations alone yield only conditional, model-based inferences rather than definitive conclusions.

Seyma Yaman Kayadibi2026-09-03

💻 computer science

MeowID: A Dual-Expert Retrieval System for Individual Cat Identification

This paper introduces MeowID, a dual-expert retrieval system that combines facial and whole-cat recognition with a face-priority framework to achieve robust individual cat identification in unconstrained environments, accompanied by the release of the new Individual Cats in the Wild (ICW) dataset.

Zhangchi Hu, Yi Shang, Haocheng Yang, Qiwei Hu, Yuzheng Li2026-09-03

💻 computer science

Similarity Finds the Fact, Not the Version: A Co-Trained Order Stamp in the Key Resolves Version Conflict in Persistent Model-Internal Memory

This paper demonstrates that adding a co-trained "order stamp" to persistent model-internal memory keys resolves version conflicts by enabling precise retrieval of specific facts based on their temporal sequence, rather than relying on capacity or recency shortcuts.

Maximiliano Rodrigo Speranza2026-09-03

💻 computer science

Modeling Loading Anomalies and Identifying Root Causes in Media Consumption Workflows on Large Social Media Platforms

This paper presents a dynamic service-dependency graph approach for modeling loading anomalies and identifying root causes in large social media platforms, achieving high accuracy in predicting delays and significantly reducing localization time compared to static methods.

Yuewei Yuan, Tianyi Xu, Jiayang Yin, Jialei Huang2026-09-03

💻 computer science

Weakly Supervised Anomaly Detection and Privacy Risk Scoring in Community Message Threads

This paper introduces RiskThread-LF, a weakly supervised and privacy-preserving framework that detects anomalous community message threads by fusing diverse behavioral signals and correcting for reporting bias, achieving superior performance over content-based and graph-based baselines while maintaining robustness under differential privacy.

Yuewei Yuan, Jiayang Yin, Jialei Huang, Tianyi Xu2026-09-03

← Previous Next →