Methodology gaps in Computer Science

Representative open questions

Showing 30 of 638 — one per source paper, highest-quality first.

• Federated learning for privacy-preserving skin cancer classification using deep neural networks (2026) · doi

  The federated learning approach using the ring topology outperforms centralized training on the ISIC dataset, but the effect of different topologies (e.g., FedAvg, FedProx) is not fully explored.

• Causal K-means clustering (2026) · doi

  The causal K-means clustering method requires separate independent samples for constructing the nuisance parameter estimator μ̂ to satisfy Assumption A2; the paper does not explore whether the sample-splitting requirement can be relaxed or how performance degrades when a single sample is used for both clustering and nuisance estimation.

• Hybrid Deep Model for Pain Intensity Classification Using Fused ECG, EMG, and GSR Signals (2026) · doi

  The study demonstrates that signal sequence ordering (GSR–ECG–EMG vs. ECG–GSR–EMG) significantly impacts hybrid BiLSTM-MHAT-CNN model performance, but lacks systematic investigation of all possible permutations and theoretical justification for optimal ordering. Future work should enumerate and evaluate all six permutations of the three physiological signals to establish evidence-based signal sequencing guidelines for multimodal pain intensity classification.

• Turbulence closure in Reynolds-averaged Navier–Stokes and flow inference around a cylinder using physics-informed neural networks and sparse experimental data (2026) · doi

  The wall-damping function for the Reynolds-force neural network is currently defined using a cubic polynomial with a fixed threshold d* = 0.1D; the sensitivity of turbulence closure predictions to alternative damping function formulations and threshold values across the Reynolds number range Re ≈ 300−300,000 has not been systematically investigated.

• AN ITERATIVE GLMM–XGBOOST ALGORITHM WITH GROUP-AWARE CONDITIONAL PERMUTATION IMPORTANCE FOR EXPLAINING MULTILEVEL ITEM RESPONSE DATA (2026) · doi

  The permutation importance analysis (Panel C, Table 2) demonstrates that standalone XGBoost's rank correlation with true feature importance degrades sharply as ICC increases (from 0.798 to 0.472 at large sample size), but the paper does not investigate the specific mechanisms causing this degradation or propose modifications to XGBoost's importance calculation for clustered data structures.

• Cross-Currency Basis Swaps Referencing Backward-Looking Rates (2026) · doi

  The pricing framework explicitly handles multi-period CCBS with tenor structure T_n, but numerical implementation details for computing the nested expectations in terms I₂ₜ, I₃ₜ, and I₄ₜ (which involve products of multiple Gamma functions with correlated parameters) are not provided. Computational complexity and convergence criteria for Monte Carlo or tree-based approximations remain unaddressed.

• FPGA-Enabled Machine Learning Applications in Earth Observation: A Systematic Review (2026) · doi

  Vision Transformers (ViTs) and recurrent neural networks (RNNs) lack mature FPGA implementations despite their superior performance in remote sensing tasks like temporal change monitoring and flight-path analysis. Current FPGA toolchains (FINN, Vitis AI) do not support transformer and recurrent architectures, including their quantized and pruned variants, creating a barrier to deploying these modern architectures on edge platforms.

• Securing IoT Devices with PUFs: Mitigating Aging and Tampering through Cryptography and Machine Learning (2026) · doi

  The linear complexity test requires M to be chosen between 500 and 5000 for the limiting distribution to provide reasonable approximation, but the paper does not specify how to select the optimal M value within this range for PUF-based cryptographic key generation in IoT devices with varying hardware constraints.

• Blockchain-integrated machine learning framework for transparent smart contract vulnerability detection (2026) · doi

  XGBoost and LightGBM showed lower recall (~0.82) compared to Random Forest and CatBoost (~0.77-0.88) on smart contract vulnerability detection due to overfitting on dominant features and sensitivity to hyperparameter tuning. The specific hyperparameter configurations that minimize this performance gap for gradient boosting methods on blockchain vulnerability datasets have not been systematically explored.

• Can deep learning-based segmentation and classification improve the detection of renal cortical abnormalities? (2026) · doi

  The paper does not investigate how CLAHE preprocessing parameters (clip limit, tile grid size) were optimized or whether these parameters are transferable across different DMSA imaging equipment and acquisition protocols. Future work should systematically evaluate CLAHE preprocessing robustness for DenseNet205 classification across DMSA systems with varying imaging parameters and reconstruction algorithms.

• Inferring High-Dimensional Dynamic Networks Changing with Multiple Covariates (2026) · doi

  The distinction between sparse edges present in all nine subgraphs (core network unaffected by covariates) versus covariate-influenced edges is determined by the tuning parameters γ1 and γ2 (set at 5×10⁻⁵ and 5×10⁻⁶), but systematic guidelines for selecting these penalty parameters in CVN estimation across different biological datasets and covariate structures are not provided.

• Umjetna inteligencija: od kritičkih promišljanja do njezine primjene u kaznenom pravosuđu (2026) · doi

  The Italian Delia algorithm processes over 1.5 million variables combining geographic crime data with witness interviews and CCTV footage, but no comparative analysis exists documenting which specific variables contribute most to prediction accuracy or how the system performs across different geographic regions and crime categories.

• Prediction of sedimentation concentration profiles in inclined suspension systems: A data-driven neural network framework (2026) · doi

  The study employed separate neural networks trained for each inclination angle to preserve regime-specific sedimentation characteristics; development of a unified ANN architecture that integrates inclination angle as an input parameter across multiple angles in inclined suspension systems has not been explored.

• Large language model based machine translation for universal multilingual understanding and translation quality enhancement (2026) · doi

  The evaluation of LLM-based machine translation systems relies on varying preprocessing pipelines, tokenization schemes, and evaluation protocols across studies, making it impossible to establish directly comparable quantitative benchmarks. A standardized evaluation framework needs to be developed that harmonizes these methodological differences for consistent cross-model comparison in large language model machine translation.

• Environmental, social, and governance sustainability: an AI-centric approach driving data standardization and automation (2026) · doi

  The social ESG category model achieved only 64% accuracy compared to 99% for environmental and governance models; the paper does not investigate why social indicator prediction substantially underperforms or propose targeted improvements to the tokenization, hyperparameter tuning, or feature engineering specifically for social KPI classification.

• Leveraging machine learning to enhance aerosol classification using Single-Particle Mass Spectrometry (2026) · doi

  The study identifies feldspar discrimination as a key challenge due to chemical overlap between Na-feldspar, K-feldspar, feldspar cSA, and feldspar cSOA subtypes. Advanced feature engineering or architectural modifications to single-particle mass spectrometry models must be developed to capture subtle spectral differences that distinguish these compositionally similar mineral species.

• Quantum Information Framework for Neural Network Generalization: A Comprehensive Experimental Analysis (2026) · doi

  The quantum analyzer computes density matrices from activation patterns using a single method (likely outer product construction); alternative density matrix formulations and their impact on von Neumann entropy and purity measurements for neural network activations remain unexplored, creating ambiguity about metric robustness.

• AI in Cybersecurity: A Systematic Review and Conceptual Audit Model (2026) · doi

  The Anti-Sheriff cybersecurity audit model is presented as a conceptual framework rather than a detailed technical implementation. The paper lacks specification of the algorithmic architecture, data input formats, and computational requirements needed to operationalize the generative and predictive AI components within the risk-calibrated assessment mechanisms.

• Large Language Models for Combinatorial Optimization: A Systematic Review (2026) · doi

  The integration of Large Language Models with automated planning and scheduling (APS) systems requires investigation of how LLMs can effectively contribute to constraint specification, plan generation, and solution validation within APS frameworks, as noted in reference [166] on prospects of incorporating LLMs in APS.

• Explainable machine learning for tracking spatial variation in leaf chlorophyll fluorescence within temperate deciduous forest canopies (2026) · doi

  While SHAP was employed to improve interpretability, the mechanistic understanding of how spectral reflectance drives chlorophyll fluorescence predictions remains limited. Future work should integrate the Random Forest model with process-based models (e.g., radiative transfer models) following the framework of Wolanin et al. (2019) to bridge predictive accuracy with photosynthetic mechanism understanding.

• Research on a strongly generalizable fault diagnosis method based on adversarial transfer learning (2026) · doi

  Conditions 12 and 13 in the target reactor exhibit highly similar multivariate trend characteristics across parameters 1, 2, and 3, making them difficult to differentiate. The HDAL adversarial transfer learning model achieved only 92.556% accuracy on these conditions due to data quality issues, requiring investigation into feature engineering or data preprocessing techniques specifically designed to enhance discriminability between similar fault conditions in cross-reactor transfer diagnosis.

• Comparative analysis of deep learning algorithms for rolling element bearing fault classification under variable loads and speeds (2026) · doi

  The study identified inconsistent class-wise performance across k-fold validation splits, with specific confusions observed between Combined Fault (CF) and Healthy Baseline (HB), as well as between Inner Race Fault (IRF) and Outer Race Fault (ORF) in certain folds. Future work should investigate class-imbalance mitigation strategies and fault-category-specific augmentation techniques for deep learning bearing fault classifiers to address these persistent misclassifications in particular fault pair combinations.

• LLM-Powered Silent Bug Fuzzing in Deep Learning Libraries via Versatile and Controlled Bug Transfer (2026) · doi

  LLM performance on LLM-Powered Self-Validation shows significant model-dependent variation in precision-recall trade-offs (Table 6): DeepSeek R1 achieves 76.92% precision but only 47.62% recall, while GPT-4.1 mini achieves 90.48% recall but only 65.52% precision. The strictness parameter governing validation behavior across different LLM models requires systematic investigation to develop model-agnostic calibration strategies for silent bug validation in deep learning fuzzing.

• Quantum-SpinalNet: a hybrid deep learning approach for mammographic breast cancer detection (2026) · doi

  The ablation study showed that QNN + SpinalNet without Swin ResUNet3+ achieved only 89.7% accuracy (Dice = 0.83), but the paper does not investigate which specific types of lesion patterns or tumor morphologies cause this performance degradation. Detailed analysis of failure modes in this configuration would clarify the necessity of transformer-based spatial encoding for particular diagnostic scenarios.

• Ocean: Object-aware Anchor-free Tracking with Matching-relation Learning (2026) · doi

  The Ocean++ tracker demonstrates performance gains on transparent object tracking in medical scenes (TOTB dataset), but the specific feature extraction and matching network mechanisms required to capture fine details of objects with weak appearance characteristics are not explicitly analyzed or generalized.

• From unstructured text to structured reasoning: a hybrid knowledge graph for Indonesian sentencing analysis (2026) · doi

  The position profile entity extraction for narcotic cases achieves only 61.3% F1-score compared to 92.7% for corruption cases; this substantial performance gap remains unexplained, with the paper attributing it to low relevance rather than investigating whether narcotic-specific organizational hierarchies require specialized entity classification or domain-adapted training data.

• Bifurcations in Lagrangian systems and geodesics I (2026) · doi

  The proof of Theorem 1.9 distinguishes cases (A), (B), and (C) based on existence of Euler-Lagrange curves αk and αλ near γ|[0,λ], but the paper does not provide explicit algorithmic or computational methods to verify which case applies for concrete Lagrangian systems with given parameter values.

• Neural Network Tools in the Arsenal of a University Teacher (2026) · doi

  The paper identifies that AI-based tools perform worse than traditional methods at precision in identifying relevant articles but better at finding unique publications, yet provides no empirical framework for measuring or optimizing the trade-off between relevance accuracy and uniqueness discovery in hybrid AI-traditional search methodologies for academic literature.

• Securing Fog-assisted IoT: An Adaptable and Efficient Threat Identification Approach (2026) · doi

  The paper does not address false positive rates and their operational impact in production fog-IoT systems, particularly how false alarms affect system performance, user trust, and resource allocation at fog nodes. Specific threshold optimization strategies and false positive mitigation techniques for DEL-based threat detection require investigation.

• A Robust Hybrid Deep Learning Model for Multiclass Depression Classification from Speech Audio (2026) · doi

  Confusion matrices revealed frequent misclassifications between Depression Stage 1 and Stage 2, particularly in models lacking attention mechanisms; future work must investigate which specific acoustic features and fine-grained feature integration strategies are needed to improve nuanced severity discrimination in multiclass depression detection.