IEEE Access

AI coding tools have come a long way, suggesting code, fixing bugs, even generating inline documentation, but full autonomy? Not just yet. A new study from researchers at Cornell, MIT CSAIL, Stanford, and UC Berkeley highlights key limitations: Managing massive codebases, understanding long-term design contexts, and making logical leaps remain hurdles for today’s AI models. The verdict: AI is best when paired with humans. Real collaboration with AI is powerful, but human insight and oversight are still essential: https://lnkd.in/eCJg2w-t 

In this featured IEEE Access article, researchers share the latest milestone in antenna engineering: a wideband, dual‑polarized 1‑bit reconfigurable reflectarray antenna designed for the Ku‑band. This innovative layout, which delivers wide bandwidth, dual‑polarization capability, and simplified fabrication, promises practical advantages for future beam‑steering applications in communications, radar, and more. Read the full article to learn more ⬇️ https://lnkd.in/ebzV5mfr

At IEEE Access, we believe innovation flourishes when passion leads the way. Whether you're exploring new technologies, conducting bold research, or championing breakthroughs, true impact blooms when you pursue what excites you.

The next giant leap in space isn’t about bigger rockets—it’s about smarter fuel strategies. Orbital refueling is gaining traction as a key enabler for longer, more sustainable missions to the Moon, Mars, and beyond. Rather than launching fully fueled rockets, spacecraft can refuel in Earth orbit, significantly extending their operational range and payload capacity. This isn’t science fiction. It’s fast becoming a reality. It also paves the way for sustainable, reusable space infrastructure. Read the full article to learn more ⬇️ https://lnkd.in/gbtJFvNR 

Excited to share that our paper has been published in IEEE Access! Title: Can We Trust AI With Our Ears? A Cross-Domain Comparative Analysis of Explainability in Audio Intelligence Read the full paper here: 🔹 ResearchGate: https://lnkd.in/gkm6Qp3B 🔹 IEEE Access: https://lnkd.in/gEYNgJH6 In this work, we explore how Explainable AI (XAI) can help us understand deep learning models in audio-based systems, from speech emotion recognition to environmental sounds and even healthcare signals like heart, lung, and cough sounds. We conducted a broad comparative study using six datasets, analyzing both audio-only and multimodal models that combine audio features with demographic data. Using XAI techniques like LIME, SHAP, and Grad-CAM, we were able to trace model predictions back to clinically and acoustically relevant features, creating more transparent, reliable, and human-aligned AI systems. I would like to express my heartfelt gratitude to Pappu Bishwas for his valuable contributions, tireless support, and insightful discussions that strengthened this research at every step. A very special thank you to Dr.Mainak Bandyopadhyay, whose exceptional guidance and mentorship throughout this journey, unwavering belief in me, and support in securing funding were all crucial in making this work possible. His encouragement and insight at every stage inspired me to push the boundaries of this research. I am also deeply grateful to Jérémie Sublime, whose collaboration was invaluable in shaping the paper, providing guidance, and helping us refine our analyses with his expertise. I would also like to acknowledge the reviewers, whose thoughtful and constructive feedback helped us improve the paper and make it stronger. Finally, a sincere thank you to KIIT - Kalinga Institute of Industrial Technology University for supporting the article processing charge and enabling the publication of this work. This study reinforces the importance of trust and explainability in AI, especially in domains where human lives and decision-making are involved. #AI #MachineLearning #DeepLearning #ExplainableAI #XAI #HealthcareAI #ArtificialIntelligence #Research #IEEEAccess #OpenAccess

I am excited to share our latest research, titled “Modeling and Predicting Uncertainty in Tidal Turbine Power Output: A Data-Driven Time-Series Approach,” published in IEEE Access. The integration of sustainable energy sources into modern power grids is crucial for achieving a more resilient and environmentally sustainable electrical infrastructure. Among these sources, tidal energy has emerged as a promising solution for expanding power generation and diversifying the renewable energy mix. However, tidal energy is influenced by variations in environmental and hydrodynamic conditions, which affect tidal turbine power output. As a result, the amount of power a tidal turbine generates at any given moment remains uncertain. This study introduces a systematic approach to quantifying, modeling, and predicting uncertainty in tidal turbine power output. Unlike conventional methods that focus on variability, this framework defines uncertainty as unpredictability and applies time-series modeling to characterize and forecast uncertainty in power generation. The findings establish a foundation for advancing tidal energy deployment and enhancing the integration of renewable resources into the power grid. The developed framework provides valuable insights for risk-based decision-making in grid management and supports the broader goal of developing resilient and sustainable electrical infrastructure. For more information, the full article is available here: https://lnkd.in/eieK-gCE    I would like to express my sincere gratitude to my advisor, Prof. Mohammad Ilbeigi, Ph.D., for his invaluable guidance and support that made this research possible. #RenewableEnergy #TidalEnergy #ElectricalInfrastructure #EnergySystems #PowerGrid #IEEEAccess #UncertaintyModeling #TimeSeriesAnalysis #SustainableEnergy #DataDrivenResearch #CleanEnergy

🚦 AI can now even manage our Indian cities traffic! Our research shows exactly how: an AI-powered framework that speeds up traffic analysis, supports urban planning, and enables smarter decision-making for Indian cities. Our research journal, "Indian Traffic Surveillance Video Summarization Using YOLO and Multi-Level Masking"(IEEE Access, Q1 Volume 13), presents a scalable AI system tailored for Indian traffic environments. By combining YOLOv8, YOLO11, Faster R-CNN, and RetinaNet with multi-level masking techniques, our framework highlights relevant vehicles and events while reducing unnecessary data for faster, focused analysis. This opens possibilities for traffic enforcement, urban planning, and anomaly detection, while maintaining flexibility for privacy and context-specific needs. It was an incredible experience collaborating with Aishwarya Gite and ASHUTOSH JOSHI and working under the guidance of Dr. Mohana, Dr.Ramakanthkumar p, K Sreelakshmi, and T. Shankar, their insights were invaluable throughout this journey. A big shoutout to the entire team for making this happen! 🚀 📄 Read here: https://lnkd.in/gnCmkhGn #TrafficTech #AI #DeepLearning #YOLO #VideoSummarization #SmartCities #UrbanMobility #ComputerVision #ResearchJournal #Innovation

In this featured article, researchers are introducing TDA‑MMRec, a ground-breaking hybrid framework that combines Topological Data Analysis (TDA) with graph learning to elevate multimodal recommendation systems (text, images, interactions). TDA‑MMRec proves that understanding data “shape” can dramatically enhance recommendation performance. Read the full article to learn more ⬇️ https://lnkd.in/eZn98pKJ