arXiv subjects

MZ Naser

Publications and source records attributed to MZ Naser.

9 recordsLinked to original sources

arXiv2026

Engineering Reasoning and Instruction (ERI) Benchmark: A Large Taxonomy-driven Dataset for Foundation Models and Agents

The Engineering Reasoning and Instruction (ERI) benchmark is a taxonomy-driven instruction dataset designed to train and evaluate engineering-capable large language models (LLMs) and agents. This dataset spans nine engineering fields (namely: civil, mechanical, electrical, chemical, environmental, aerospace, materials, fire, and industrial engineering) and 55 subdomains, and is crossed with seven intent types (i.e., definition, explanation, calculation, comparison, design/synthesis, troubleshooting, and code-related) and three difficulty tiers (undergraduate, graduate, and professional), yielding 57,750 records with field/subdomain/type/difficulty metadata and solution formatting. We examined ERI via seven LLMs and report a statistically significant three-tier performance structure, with frontier models (GPT-5, Claude Sonnet 4, DeepSeek V3.1) achieving mean scores above 4.30 on a five-point scale, while mid-tier and smaller models exhibited progressively higher failure rates and steeper performance degradation on graduate-level questions. To address circularity concerns inherent in LLM benchmarks, we developed a convergent validation protocol that leverages cross-provider independence, multi-judge averaging, and frontier-model agreement analysis to empirically bound hallucination risk to 1.7%. ERI is released with taxonomy specifications, validation scripts, and an evaluation harness to enable reproducible comparisons and regression testing for instruction tuning, routing, retrieval-augmented evaluation, and agentic tool-use workflows in engineering settings.

cs.AI↗

arXiv2026

How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations

Large language models (LLMs) have been noted to fabricate scholarly citations, yet the scope of this behavior across providers, domains, and prompting conditions remains poorly quantified. We present one of the largest citation hallucination audits to date, in which 10 commercially deployed LLMs were prompted across four academic domains, generating 69,557 citation instances verified against three scholarly databases (namely, CrossRef, OpenAlex, and Semantic Scholar). Our results show that the observed hallucination rates span a fivefold range (between 11.4% and 56.8%) and are strongly shaped by model, domain, and prompt framing. Our results also show that no model spontaneously generates citations when unprompted, which seems to establish hallucination as prompt-induced rather than intrinsic. We identify two practical filters: 1) multi-model consensus (with more than 3 LLMs citing the same work yields 95.6% accuracy, a 5.8-fold improvement), and 2) within-prompt repetition (with more than 2 replications yields 88.9% accuracy). In addition, we present findings on generational model tracking, which reveal that improvements are not guaranteed when deploying newer LLMs, and on capacity scaling, which appears to reduce hallucination within model families. Finally, a lightweight classifier trained solely on bibliographic string features is developed to classify hallucinated citations from verified citations, achieving AUC 0.876 in cross-validation and 0.834 in LOMO generalization (without querying any external database). This classifier offers a pre-screening tool deployable at inference time.

cs.CL↗

arXiv2025

Philosophy-informed Machine Learning

Philosophy-informed machine learning (PhIML) directly infuses core ideas from analytic philosophy into ML model architectures, objectives, and evaluation protocols. Therefore, PhIML promises new capabilities through models that respect philosophical concepts and values by design. From this lens, this paper reviews conceptual foundations to demonstrate philosophical gains and alignment. In addition, we present case studies on how ML users/designers can adopt PhIML as an agnostic post-hoc tool or intrinsically build it into ML model architectures. Finally, this paper sheds light on open technical barriers alongside philosophical, practical, and governance challenges and outlines a research roadmap toward safe, philosophy-aware, and ethically responsible PhIML.

cs.AI↗

arXiv2025

A Look into How Machine Learning is Reshaping Engineering Models: the Rise of Analysis Paralysis, Optimal yet Infeasible Solutions, and the Inevitable Rashomon Paradox

The widespread acceptance of empirically derived codal provisions and equations in civil engineering stands in stark contrast to the skepticism facing machine learning (ML) models, despite their shared statistical foundations. This paper examines this philosophical tension through the lens of structural engineering and explores how integrating ML challenges traditional engineering philosophies and professional identities. Recent efforts have documented how ML enhances predictive accuracy, optimizes designs, and analyzes complex behaviors. However, one might also raise concerns about the diminishing role of human intuition and the interpretability of algorithms. To showcase this rarely explored front, this paper presents how ML can be successfully integrated into various engineering problems by means of formulation via deduction, induction, and abduction. Then, this paper identifies three principal paradoxes that could arise when adopting ML: analysis paralysis (increased prediction accuracy leading to a reduced understanding of physical mechanisms), infeasible solutions (optimization resulting in unconventional designs that challenge engineering intuition), and the Rashomon effect (where contradictions in explainability methods and physics arise). This paper concludes by addressing these paradoxes and arguing the need to rethink epistemological shifts in engineering and engineering education and methodologies to harmonize traditional principles with ML.

cs.LG↗

arXiv2024

SPINEX_ Symbolic Regression: Similarity-based Symbolic Regression with Explainable Neighbors Exploration

This article introduces a new symbolic regression algorithm based on the SPINEX (Similarity-based Predictions with Explainable Neighbors Exploration) family. This new algorithm (SPINEX_SymbolicRegression) adopts a similarity-based approach to identifying high-merit expressions that satisfy accuracy- and structural similarity metrics. We conducted extensive benchmarking tests comparing SPINEX_SymbolicRegression to over 180 mathematical benchmarking functions from international problem sets that span randomly generated expressions and those based on real physical phenomena. Then, we evaluated the performance of the proposed algorithm in terms of accuracy, expression similarity in terms of presence operators and variables (as compared to the actual expressions), population size, and number of generations at convergence. The results indicate that SPINEX_SymbolicRegression consistently performs well and can, in some instances, outperform leading algorithms. In addition, the algorithm's explainability capabilities are highlighted through in-depth experiments.

cs.LG↗

arXiv2024

SPINEX-Optimization: Similarity-based Predictions with Explainable Neighbors Exploration for Single, Multiple, and Many Objectives Optimization

This article introduces an expansion within SPINEX (Similarity-based Predictions with Explainable Neighbors Exploration) suite, now extended to single, multiple, and many objective optimization problems. The newly developed SPINEX-Optimization algorithm incorporates a nuanced approach to optimization in low and high dimensions by accounting for similarity across various solutions. We conducted extensive benchmarking tests comparing SPINEX-Optimization against ten single and eight multi/many optimization algorithms over 55 mathematical benchmarking functions and realistic scenarios. Then, we evaluated the performance of the proposed algorithm in terms of scalability and computational efficiency across low and high dimensions, number of objectives, and population sizes. The results indicate that SPINEX-Optimization consistently outperforms most algorithms and excels in managing complex scenarios, especially in high dimensions. The algorithm's capabilities in explainability, Pareto efficiency, and moderate complexity are highlighted through in-depth experiments and visualization methods.

stat.AP↗

arXiv2024

SPINEX-TimeSeries: Similarity-based Predictions with Explainable Neighbors Exploration for Time Series and Forecasting Problems

This paper introduces a new addition to the SPINEX (Similarity-based Predictions with Explainable Neighbors Exploration) family, tailored specifically for time series and forecasting analysis. This new algorithm leverages the concept of similarity and higher-order temporal interactions across multiple time scales to enhance predictive accuracy and interpretability in forecasting. To evaluate the effectiveness of SPINEX, we present comprehensive benchmarking experiments comparing it against 18 algorithms and across 49 synthetic and real datasets characterized by varying trends, seasonality, and noise levels. Our performance assessment focused on forecasting accuracy and computational efficiency. Our findings reveal that SPINEX consistently ranks among the top 5 performers in forecasting precision and has a superior ability to handle complex temporal dynamics compared to commonly adopted algorithms. Moreover, the algorithm's explainability features, Pareto efficiency, and medium complexity (on the order of O(log n)) are demonstrated through detailed visualizations to enhance the prediction and decision-making process. We note that integrating similarity-based concepts opens new avenues for research in predictive analytics, promising more accurate and transparent decision making.

stat.ME↗

arXiv2024

SPINEX-Clustering: Similarity-based Predictions with Explainable Neighbors Exploration for Clustering Problems

This paper presents a novel clustering algorithm from the SPINEX (Similarity-based Predictions with Explainable Neighbors Exploration) algorithmic family. The newly proposed clustering variant leverages the concept of similarity and higher-order interactions across multiple subspaces to group data into clusters. To showcase the merit of SPINEX, a thorough set of benchmarking experiments was carried out against 13 algorithms, namely, Affinity Propagation, Agglomerative, Birch, DBSCAN, Gaussian Mixture, HDBSCAN, K-Means, KMedoids, Mean Shift, MiniBatch K-Means, OPTICS, Spectral Clustering, and Ward Hierarchical. Then, the performance of all algorithms was examined across 51 synthetic and real datasets from various domains, dimensions, and complexities. Furthermore, we present a companion complexity analysis to compare the complexity of SPINEX to that of the aforementioned algorithms. Our results demonstrate that SPINEX can outperform commonly adopted clustering algorithms by ranking within the top-5 best performing algorithms and has moderate complexity. Finally, a demonstration of the explainability capabilities of SPINEX, along with future research needs, is presented.

cs.LG↗

arXiv2024

SPINEX: Similarity-based Predictions with Explainable Neighbors Exploration for Anomaly and Outlier Detection

This paper presents a novel anomaly and outlier detection algorithm from the SPINEX (Similarity-based Predictions with Explainable Neighbors Exploration) family. This algorithm leverages the concept of similarity and higher-order interactions across multiple subspaces to identify outliers. A comprehensive set of experiments was conducted to evaluate the performance of SPINEX. This algorithm was examined against 21 commonly used anomaly detection algorithms, namely, namely, Angle-Based Outlier Detection (ABOD), Connectivity-Based Outlier Factor (COF), Copula-Based Outlier Detection (COPOD), ECOD, Elliptic Envelope (EE), Feature Bagging with KNN, Gaussian Mixture Models (GMM), Histogram-based Outlier Score (HBOS), Isolation Forest (IF), Isolation Neural Network Ensemble (INNE), Kernel Density Estimation (KDE), K-Nearest Neighbors (KNN), Lightweight Online Detector of Anomalies (LODA), Linear Model Deviation-based Detector (LMDD), Local Outlier Factor (LOF), Minimum Covariance Determinant (MCD), One-Class SVM (OCSVM), Quadratic MCD (QMCD), Robust Covariance (RC), Stochastic Outlier Selection (SOS), and Subspace Outlier Detection (SOD), and across 39 synthetic and real datasets from various domains and of a variety of dimensions and complexities. Furthermore, a complexity analysis was carried out to examine the complexity of the proposed algorithm. Our results demonstrate that SPINEX achieves superior performance, outperforms commonly used anomaly detection algorithms, and has moderate complexity (e.g., O(n log n d)). More specifically, SPINEX was found to rank at the top of algorithms on the synthetic datasets and the 7th on the real datasets. Finally, a demonstration of the explainability capabilities of SPINEX, along with future research needs, is presented.

cs.LG↗