About the Journal

The Journal of Explainable Artificial Intelligence (JXAI) is a peer-reviewed, open-access scientific journal published online by CV AIR PUBLISHING. First published in 2026, JXAI is issued biannually (two issues per year, in June and December) and is freely accessible to readers worldwide without subscription or access fee.

JXAI is dedicated to advancing research at the intersection of Explainable Artificial Intelligence (XAI) and AI Optimization. The journal publishes high-quality theoretical and applied research that enhances the transparency, interpretability, trustworthiness, and performance of AI and machine learning systems. JXAI serves as a global platform for researchers, industry practitioners, regulators, and policymakers to share findings that bridge the gap between model performance optimization and human-understandable AI.

The journal accepts manuscripts exploring methods, theories, algorithms, software, and real-world case studies across the full AI model lifecycle from architecture design and hyperparameter optimization to deployment, monitoring, auditing, and explainability of decisions made by AI systems.

─────────────────────────────────────
FOCUS AND SCOPE
─────────────────────────────────────

Topics covered include, but are not limited to:

[A] EXPLAINABILITY & INTERPRETABILITY
• Feature attribution methods: SHAP, LIME, Integrated Gradients, saliency maps
• Surrogate models, rule/decision-set extraction, counterfactual & contrastive explanations
• Concept-based, prototype-based, and attention-based explanations
• Causal and mechanistic interpretability of neural networks
• Ante-hoc (glass-box) and post-hoc explanation of black-box models
• XAI for LLMs, transformers, foundation models, and generative AI
• Theory and guarantees: faithfulness, identifiability, calibration in explanations
• Robustness, security, and privacy of explanations

[B] AI OPTIMIZATION
• Hyperparameter optimization: Bayesian optimization, evolutionary algorithms, particle swarm, grid/random search
• Neural Architecture Search (NAS) and AutoML
• Multi-objective optimization for accuracy–interpretability–latency–privacy trade-offs
• Optimization for deep learning, ensemble methods, and federated learning
• Scalable and efficient training pipelines for large-scale ML models

[C] HUMAN-CENTERED & TRUSTWORTHY AI
• Human-AI collaboration, explanation interfaces, cognitive load, and user studies
• Trust, fairness, accountability, and transparency (algorithmic recourse, bias auditing)
• Regulatory compliance: EU AI Act, GDPR, and emerging AI governance frameworks
• Ethical, societal, and environmental impact of AI systems

[D] EVALUATION, TOOLS & APPLICATIONS
• Benchmarks, reproducibility, and evaluation protocols for XAI and AI optimization
• Libraries, frameworks, visualization toolkits, and MLOps practices
• Real-world applications: healthcare, finance, legal systems, autonomous systems, cybersecurity, NLP, time series, and scientific discovery

─────────────────────────────────────
PUBLICATION INFORMATION
─────────────────────────────────────

  • Publisher : CV AIR PUBLISHING
  • Media : Online (Electronic)
  • Access : Open Access
  • First Published : 2026
  • Frequency : Biannual (June & December)
  • Peer Review : Double-blind
  • Language : English (optional translated abstracts)
  • Article Types : Research articles, short papers/letters, surveys/tutorials,
    systems & software papers, application/case study papers,
    reproducibility reports

─────────────────────────────────────
MISSION STATEMENT
─────────────────────────────────────

JXAI's mission is to advance rigorous, human-centered, and reproducible research at the intersection of AI explainability and AI optimization, so that AI and ML systems are simultaneously more performant, transparent, trustworthy, fair, and accountable to the people and societies they affect.